DETAILED ACTION
This final Office action is responsive to amendments filed December 16th, 2025. Claims 1, 3-7, 9, 11, 13, 14, 16-18, 21-24, 26, 28, 30, 31, 33-36, and 38 have been amended. Claims 1-38 are presented for examination.
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Priority
Receipt is acknowledged of certified copies of papers required by 37 CFR 1.55.
Response to Arguments
Applicant’s arguments, see page 15, filed 12/16/25, with respect to claims 3, 16, 17, 22, 33, 34, and 38 have been fully considered and are persuasive. The objection of 9/23/25 has been withdrawn.
Applicant's arguments regarding claim rejections under 35 USC 101 filed 12/16/25 have been fully considered but they are not persuasive.
On pages 15-21 of the provided remarks, Applicant argues that the amended claims present statutory subject matter. Beginning on page 16 of the provided remarks, Applicant argues “this exemplary claimed system, in amended claim 1, cannot be said to merely recite “certain methods of organizing human activity and mental processes” because as further amended now clearly recites sufficient elements to clearly recite that, even assuming a judicial exception is still recited in the claim limitations, any such exception is now clearly integrated into a practical application as required by the 2019 PEG.” While Applicant argues on pages 16-17 that the claimed invention is directed to “a system for developing a bi-directional cloud based unified digital platform based virtual power banks which are power entities hosted in the unified digital platform” Examiner asserts that the claims recitation of generating virtual power banks and generating dynamic actionable items relating to one or more operational parameters of the virtual power bank are managing personal behavior as the interactions within the unified digital platform occur between “an end- consumer and a utility retailer” per paragraph [0018] of the as-filed Specification. Therefore, the claims recite Certain Methods of Organizing Human Activity as well as Mental Processes. Applicant’s arguments are not persuasive.
Further, on page 17 of the provided remarks, Applicant argues “the aforementioned features of generating and hosting power entities (virtual power banks) in a unified digital platform and optimizing the generated virtual power banks through a specific way of iterative training and optimization of generated models, in the manner recited in amended claim 1, cannot be said to be abstract ideas of certain methods of organizing human activity and mental processes.” Examiner respectfully disagrees and asserts, as stated above, that the generation and hosting of power entities (virtual power banks” in a unified digital platform is directed to managing personal behavior between an end consumer and a utility retailer. Additionally, while Applicant argues that the amended iterative training and optimization of generated models” is performed in “a specific way” the amended claims recite “wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively” which does not actively recite the iterative training of generated models as argued. Additionally, the argued optimization method of “gradient descent optimization technique and a stochastic gradient descent optimization technique” recite the abstract idea of mathematical operations in the form of mathematical calculations as the optimization methods are known mathematical forms of optimization. Therefore, the claims recite an abstract idea. Applicant’s arguments are not persuasive.
Continuing on pages 17-18 of the provided remarks, Applicant argues “amended claim 1 provides a concrete, useful and tangible result of a unified cloud-based digital platform hosting generated virtual power banks which provides for efficient management of power generated from renewable energy sources”. Examiner respectfully disagrees and asserts that while the claim recites, “generate virtual power banks by employing the first data type and the second data type fetched from the database, wherein the virtual power banks are power entities hosted in the unified digital platform” the high-level recitation of the generation of the power banks is equivalent to retrieving first and second data from various databases to establish a power entity platform. Per cited paragraph [0018], “The system 100 provides for power/energy management and settlement (e.g., purchase and sale of renewable energy) in the form of the virtual power banks through the unified digital platform 106a.” This is a commercial based function that recites the abstract idea of Certain Methods of Organizing Human Activity. Further, while Applicant argues that the system “provides for efficient management of power”, Examiner cites, “[M]erely adding computer functionality to increase the speed or efficiency of the process does not confer patent eligibility on an otherwise abstract idea”); 2019 Revised Guidance at 55. See also Trading Techs. Int’l, Inc. v. IBG LLC, 921 F.3d 1084, 1090 (Fed. Cir. 2019)”. Therefore, Applicant’s argument is not persuasive.
On page 18 of the provided remarks, Applicant cites PTAB (Appeal 2018-007443 – Decided October 10, 2019, referred to herein as the ‘443 Appeal) to argue “Applicant’s processor qualifies as a special purpose computer and cannot be said to be generic computer component that falls within the mental processes grouping of an abstract idea.” Examiner begins by asserting that the argued “train and generate a Machine Learning (ML) model iteratively for optimizing computed initial weightage values for each sub-variable associated with variables that correspond dynamic actionable items of the generated virtual banks, and comparing results in every iteration by changing hyperparameters until an accurate model with less error rate is generated for generating optimized virtual banks” is an amended limitation not presently included with the mental process abstract idea grouping. The amended claim recites, “compute an optimized final weightage value of the generated virtual power banks, accessed via a the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables”, which are mental observations and evaluations of the human mind. While Applicant argues the analysis of ‘443 Appeal, it is unclear where Applicant is determining the argued decision distinguishes “a special purpose computer and cannot be said to be generic computer component”. Per Examiner’s review of the argued decision regarding the application directed to applying AI classification technologies and combinational logic to predict whether machines need to be serviced, and whether there is likely to be equipment failure in a system, regarding the argued mental process argument, the PTAB ruled “Specifically, we do not find ‘monitoring the operation of machines,’ as recited in the instant application, is a fundamental economic principle (such as hedging, insurance, or mitigating risk). Rather, the claims recite monitoring operation of machines using neural networks, logic decision trees, confidence assessments, fuzzy logic, smart agent profiling, and case-based reasoning.” However, this logic is not analogous to the single limitation’s recitation of “wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively” within the present claim. Therefore, Applicant’s arguments are not persuasive.
Applicant continues on pages 18-19 of the provided remarks to argue “developing of the unified cloud-based digital platform hosting the generated virtual power banks, as recited in amended claim 1, is computationally complex and cannot be performed mentally.” While Applicant cites to various claim limitations, Examiner asserts as stated below, that the identified limitations as reciting mental processes include, “analyze and process the one or more record types for generating a second data type, wherein the second data type is stored in the database; generate one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type; compute an optimized final weightage value of the generated virtual power banks, accessed via a the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique”. Therefore, arguments regarding additional limitations are moot. While Applicant argues, regarding the second data, that “the second data type is diverse in natures and its generation is directly linked to the record types (i.e., the first data type) which is received from multiple sources that entail complex data processing and therefore, cannot be said to be a mere method of organizing human activity and mental process” Examiner asserts that the claim merely recites, “analyze and process the one or more record types for generating a second data type, wherein the second data type is stored in the database” which when recited at a high-level covers evaluation and judgment of the human mind. Citing paragraphs [0024] and [0026], Applicant argues that this “entails processing a specific combination of data types and therefore, cannot be said to be a mere method of organizing human activity and mental process”. Examiner respectfully disagrees and asserts that the claim recites, “generate one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type” which are recited with a high-level of generality such that the generation covers evaluation and judgment of the human mind. Additionally, the “generated from the first data type that includes data related to real-time power supply and demand received from one or more utilities and data related to multiple variables, and the second data type fetched from the database” and listed types of action items found within the cited paragraphs does not support the argued “processing a specific combination of data types”. Therefore, the claims recite an abstract idea. Applicant’s arguments are not persuasive.
On page 19 of the provided remarks, Applicant argues “the claims have been amended to clearly demonstrate the specific unconventional technological solution to the identified technological problem of existing interfaces which does not allow end-consumers to access renewable energy sources for consumption and existing digital platforms which does not provide functionalities to visualize power sources which may be suitably utilized and re-utilized”. Examiner respectfully disagrees and asserts that the claimed “render a Graphical User Interface (GUI) on a user device for providing visualization functionalities related to virtual power banks for segregating power generated from renewable energy sources into a quantum based on demand on the unified digital platform” does not present “end-consumers to access renewable energy sources for consumption and existing digital platforms”. Per MPEP 2106.05(a) “The full scope of the claim under the BRI should be considered to determine if the claim reflects an improvement in technology (e.g., the improvement described in the specification).” Examiner asserts that the claim does not present the argued improvement in technology. The 35 USC 101 rejection is maintained. Applicant’s arguments are not persuasive.
Applicant’s arguments, see pages 21-26, filed 12/16/25, with respect to the rejection(s) of claim(s) 1-6, 11-13, 21-24, 28-30, and 38 under 35 USC 102 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1).
Applicant’s arguments, see pages 25-26, filed 12/16/25, with respect to the rejection(s) of claim(s) 7-10, 14-20, 25-27, and 31-37 under 35 USC 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1) in view of Forbes, Jr. (U.S 2017/0358041 A1) regarding claims 7-10 and 25-27; Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1) in view of Willcock (US 2022/0083854 A1) regarding claims 14 and 31; Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1) in view of Bright (U.S 2018/0294649 A1) regarding claims 15-20; and Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1) in view of Willcock (US 2022/0083854 A1) in view of Bright (U.S 2018/0294649 A1) regarding claims 32-37.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1-38 are rejected under 35 U.S.C. 101 because the claimed invention is directed to non-statutory subject matter;
When considering subject matter eligibility under 35 U.S.C. 101, it must be determined whether the claim is directed to one of the four statutory categories of invention, i.e., process, machine, manufacture, or composition of matter. If the claim does fall within one of the statutory categories, it must then be determined whether the claim is directed to a judicial exception (i.e., law of nature, natural phenomenon, and abstract idea), and if so, it must additionally be determined whether the claim is a patent-eligible application of the exception. If an abstract idea is present in the claim, any element or combination of elements in the claim must be sufficient to ensure that the claim amounts to significantly more than the abstract idea itself.
Step 1: Independent claims 1 (system), 23 (method), 38 (computer program product) and dependent claims 2-22 and 23-37, respectively, fall within at least one of the four statutory categories of 35 U.S.C. 101: (i) process; (ii) machine; (iii) manufacture; or (iv) composition of matter. Claim 1 is directed to a system (i.e. machine), claim 23 is directed to a method (i.e. process), and claim 38 is directed to a computer program product (i.e., manufacture).
Step 2A Prong 1: The independent claims recite developing a bi-directional cloud based unified digital platform based virtual power banks, the method is implemented by a processor configured to execute instructions stored in a memory, the method comprises: receiving a first data type from multiple sources and store the first data type in a database as one or more record types associated with each of the multiple sources; analyzing and processing the one or more record types for generating a second data type, wherein the second data type is stored in the database; generating virtual power banks by employing the first data type and the second data type fetched from the database, wherein the virtual power banks are power entities hosted in the unified digital platform, and wherein the virtual power banks are associated with one or more attributes; generating one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type; computing an optimized final weightage value of the generated virtual power banks, accessed via the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub-variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables, and wherein the computed initial weightage values for each of the sub-variables is optimized by training a model iteratively, and wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique; and rendering a Graphical User Interface (GUI) on a user device for providing visualization functionalities related to virtual power banks for segregating power generated from renewable energy sources into a quantum based on demand on the unified digital platform (Certain Method of Organizing Human Activity, Mental Process, and Mathematical Concepts), which are considered to be abstract ideas (See PEG 2019 and MPEP 2106.05). [Examiner notes the underlined limitations above recite the abstract idea].
The steps/functions disclosed above and in the independent claims recite the abstract idea of Certain Methods of Organizing Human Activity because the claimed limitations are generating virtual power banks and generating dynamic actionable items relating to operational parameters of the virtual power banks, which is managing personal behavior. The Applicant’s claimed limitations are generating virtual power banks and generating dynamic actionable items relating to operational parameters of the virtual power banks, which recite the abstract idea of Organizing Human Activity.
The steps/functions disclosed above and in the independent claims recite the abstract idea of Mental Process because the claimed limitations are analyzing and processing the one or more record types for generating a second data type, wherein the second data type is stored in the database; generating one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type; computing an optimized final weightage value of the generated virtual power banks, accessed via a the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique, which are functions of the human mind in the form of observation, judgment, and evaluation. The Applicant’s claimed limitations are analyze and process the one or more record types for generating a second data type, wherein the second data type is stored in the database; generate one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type; compute an optimized final weightage value of the generated virtual power banks, accessed via a the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique, which recite the abstract idea of Mental Process.
The steps/functions disclosed above and in the independent claims recite the abstract idea of Mathematical Concepts because the claimed limitations are generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique, which are mathematical calculations. The Applicant’s claimed limitations are optimizing models utilizing a gradient descent optimization technique and a stochastic gradient descent optimization technique, which recite the abstract idea of Mathematical Concepts.
Dependent claims 14 and 31 recite the abstract idea of Mathematical Concepts because the claimed limitations are carrying out a first optimization operation for optimizing the computer initial weightage values for each of the sub-variables based on various optimization techniques including: gradient descent optimization technique; stochastic gradient descent optimization technique; an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa; a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants, which are mathematical calculations. The Applicant’s claimed limitations are performing optimization operations for computing the initial weightage values of sub-variables, which recite the abstract idea of Mathematical Concepts.
Dependent claim 17-20 and 34-37 recite the abstract idea of Mathematical Concepts because the claimed limitations are carrying out a second optimization operation for optimizing the initial weightage values by computing an average of the numerical values assigned to each of the sub-variable in either of each row or column of the matrix to generate a first weightage value for each of the variables; the virtual power bank generation unit carries out a third optimization operation by computing an average of the final weightage values for each of the variables associated with the dynamic actionable items for each row of a matrix; determines a maximum average value from the computed final weightage average values associated with all the dynamic actionable items across each column of the matrix to determine a second weightage value; and determines the operational parameters for the virtual power banks by carrying out a multiplication operation between the computed second weightage value and the threshold value, and then adjusting against the base value, which are mathematical calculations. The Applicant’s claimed limitations are performing optimization operations based on the values associated with the dynamic actionable items computing averages & performing multiplication of associated values, which recite the abstract idea of Mathematical Concepts.
In addition, dependent claims 2-6, 8, 11-13, 15-16, 22, 24-25, 28-30, and 32-33 further narrow the abstract idea and recite further defining the first data types, record types, deriving second data types, the attributes, operation of virtual power banks using smart contracts, dynamic actionable items, analyzing interdependency between sub-variables, assigning labels to sub-variables, tracing the renewable energy source to generate the power obtainable, and determine unused power. These processes are similar to the abstract idea noted in the independent claims because they further the limitations of the independent claims which recite a certain method of organizing human activity which include commercial interactions such as business relations; managing interactions; and mental processes. Accordingly, these claim elements do not serve to confer subject matter eligibility to the claims since they recite abstract ideas. Dependent claims 7, 9-10, 21, and 26-27 will be discussed in Prong 2 analysis below.
Step 2A Prong 2: In this application, the above “receiving a first data type from multiple sources and store the first data type in a database as one or more record types associated with each of the multiple sources; wherein the second data type is stored in the database; employing the first data type and the second data type fetched from the database” steps/functions of the independent claims would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because receiving/storing data and displaying data merely add insignificant extra-solution activity and merely adds the words to apply it with the judicial exception. Also, the claimed “A system for developing a bi-directional cloud based unified digital platform based virtual power banks, the system comprises: a memory storing program instructions; and a processor executing program instructions stored in the memory; a database; a Graphical User Interface (GUI) on a user device; a data analytics unit executed by the processor; a cloud platform; a virtual power bank generation unit executed by the processor; a visualization unit executed by the processor; a Graphical User Interface (GUI); a user device; a processor configured to execute instructions stored in a memory; a database; A computer program product comprises: a non-transitory computer-readable medium having computer program code stored thereon, the computer-readable program code comprising instructions” would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
Independent claims 1, 23, and 38 recite “wherein the computed initial weightage values for each of the sub-variables is optimized by training a model iteratively”. The “by training a model iteratively” are recited so generically (no details whatsoever are provided other than that they are general purpose computing components) that they represent no more than mere instructions to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. These limitations would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
Dependent claims 9 and 26 recite “the first type of smart contract has a unique hash key function”. The “unique hash key function” are recited so generically (no details whatsoever are provided other than that they are general purpose computing components) that they represent no more than mere instructions to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. These limitations would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
Dependent claims 10 and 27 recite, “wherein the first type of smart contract is bifurcated into multiple sub-contracts which are associated with the virtual power banks and are hosted in a cloud platform”. The “bifurcated into multiple sub-contracts” are recited so generically (no details whatsoever are provided other than that they are general purpose computing components) that they represent no more than mere instructions to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. These limitations would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
Dependent claims 14 and 31 recite the following limitation, “employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation.” The “train a model iteratively that results in a maximum and minimum function evaluation” are recited so generically (no details whatsoever are provided other than that they are general purpose computing components) that they represent no more than mere instructions to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. These limitations would not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
In addition, dependent claims 2-6, 8, 11-16, 22, 24-25, 28-33 further narrow the abstract idea and dependent claims 15, 21, and 32 additionally recite “captures data associated with the sub-variables for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks”, “providing visualization functionalities comprising a dashboard, an application, and a portal for end-consumers to access the unified digital platform for operating or generating the virtual power banks”, and “data associated with the sub- variables is captured for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks” which do not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because receiving/storing data and displaying data merely add insignificant extra-solution activity and the claimed “virtual power bank generation unit executed by the processor” and “a visualization unit executed by the processor and configured to render a Graphical User Interface (GUI) on a user device” which do not account for additional elements that integrate the judicial exception (e.g. abstract idea) into a practical application because the claimed structure merely adds the words to apply it with the judicial exception and mere instructions to implement an abstract idea on a computer (See PEG 2019 and MPEP 2106.05).
The claimed “A system for developing a bi-directional cloud based unified digital platform based virtual power banks, the system comprises: a memory storing program instructions; and a processor executing program instructions stored in the memory; a database; a Graphical User Interface (GUI) on a user device; a data analytics unit executed by the processor; a cloud platform; a virtual power bank generation unit executed by the processor; a visualization unit executed by the processor; a Graphical User Interface (GUI); a user device; a processor configured to execute instructions stored in a memory; a database; A computer program product comprises: a non-transitory computer-readable medium having computer program code stored thereon, the computer-readable program code comprising instructions” are recited so generically (no details whatsoever are provided other than that they are general purpose computing components and regular office supplies) that they represent no more than mere instructions to apply the judicial exception on a computer. These limitations can also be viewed as nothing more than an attempt to generally link the use of the judicial exception to the technological environment of a computer. Even when viewed in combination, the additional elements in the claims do no more than use the computer components as a tool. There is no change to the computers and other technology that is recited in the claim, and thus the claims do not improve computer functionality or other technology (See PEG 2019).
Step 2B: When analyzing the additional element(s) and/or combination of elements in the claim(s) other than the abstract idea per se the claim limitations amount(s) to no more than: a general link of the use of an abstract idea to a particular technological environment and merely amounts to the application or instructions to apply the abstract idea on a computer (See MPEP 2106.05 and PEG 2019). Further, method claims 23-37; system claims 1-22; and computer program product claim 38 recite “A system for developing a bi-directional cloud based unified digital platform based virtual power banks, the system comprises: a memory storing program instructions; and a processor executing program instructions stored in the memory; a database; a Graphical User Interface (GUI) on a user device; a data analytics unit executed by the processor; a cloud platform; a virtual power bank generation unit executed by the processor; a visualization unit executed by the processor; a Graphical User Interface (GUI); a user device; a processor configured to execute instructions stored in a memory; a database; A computer program product comprises: a non-transitory computer-readable medium having computer program code stored thereon, the computer-readable program code comprising instructions”; however, these elements merely facilitate the claimed functions at a high level of generality and they perform conventional functions and are considered to be general purpose computer components which is supported by Applicant’s specification in Paragraphs 0047 and 0049 and Figures 1 & 4. The Applicant’s claimed additional elements are mere instructions to implement the abstract idea on a general purpose computer and generally link of the use of an abstract idea to a particular technological environment. Also, the above “receiving a first data type from multiple sources and store the first data type in a database as one or more record types associated with each of the multiple sources; wherein the second data type is stored in the database; employing the first data type and the second data type fetched from the database” steps/functions of the independent claims would not account for significantly more than the abstract idea because receiving data and displaying/presenting data (See MPEP 2106.05) have been identified as well-known, routine, and conventional steps/functions to one of ordinary skill in the art. When viewed as a whole, these additional claim element(s) do not provide meaningful limitation(s) to transform the abstract idea into a patent eligible application of the abstract idea such that the claim(s) amounts to significantly more than the abstract idea itself.
Next, when the “unique hash key function” is evaluated as an additional element, this feature is recited at a high level of generality and encompasses well-understood, routine, and conventional prior art activity. See, e.g., Suga et al. US 2008/0152133, noting in paragraph [0088] that “an algorithm for calculating a node-keys-assigning chart M (G) representing key generation uses the conventional technique of the hash key generating method.” Accordingly, the use of unique hash key functions does not add significantly more to the claim.
Next, when the “bifurcation of smart contracts” is evaluated as an additional element, this feature is recited at a high level of generality and encompasses well-understood, routine, and conventional prior art activity. See, e.g., Mandal et al. US 2023/0006835, noting in paragraph [0013] that “Furthermore, conventional blockchain integration systems may attempt to rely on an additional, separate blockchain and/or may employ identity proofs that rely on substantial overhead to execute.” Accordingly, the use of bifurcation of smart contracts does not add significantly more to the claim.
Next, when the “machine learning” is evaluated as an additional element, this feature is recited at a high level of generality and encompasses well-understood, routine, and conventional prior art activity. See, e.g., Balsiger et al., US 2012/0054642, noting in paragraph [0077] that “Machine learning is well known to those skilled in the art.” See also, Djordjevic et al. US 2013/0018651, noting in paragraph [0019] that “As known in the art, a generative model can be used in machine learning to model observed data directly.” See also, Bauer et al., US 2017/0147941, noting at paragraph [0002] that “Problems of understanding the behavior or decisions made by machine learning models have been recognized in the conventional art and various techniques have been developed to provide solutions.” Accordingly, the use of machine learning to generate a learning model does not add significantly more to the claim.
In addition, claims 2-6, 8, 11-16, 22, 24-25, 28-33 further narrow the abstract idea identified in the independent claims. The Examiner notes that the dependent claims merely further define the data being analyzed and how the data is being analyzed. Similarly, claims 15, 21, and 32 additionally recite “captures data associated with the sub-variables for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks”, “providing visualization functionalities comprising a dashboard, an application, and a portal for end-consumers to access the unified digital platform for operating or generating the virtual power banks”, and “data associated with the sub- variables is captured for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks” which do not account for additional elements that amount to significantly more than the abstract idea because receiving data and displaying/presenting data (See MPEP 2106.05) have been identified as well-known, routine, and conventional steps/functions to one of ordinary skill in the art and the claimed “virtual power bank generation unit executed by the processor” and “a visualization unit executed by the processor and configured to render a Graphical User Interface (GUI) on a user device” which do not account for additional elements that amount to significantly more than the abstract idea because the claimed structure merely amounts to the application or instructions to apply the abstract idea on a computer and does not move beyond a general link of the use of an abstract idea to a particular technological environment (See MPEP 2106.05). The additional limitations of the independent and dependent claim(s) when considered individually and as an ordered combination do not amount to significantly more than the abstract idea. The examiner has considered the dependent claims in a full analysis including the additional limitations individually and in combination as analyzed in the independent claim(s). Therefore, the claim(s) are rejected under 35 U.S.C. 101 as being directed to non-statutory subject matter.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
This application currently names joint inventors. In considering patentability of the claims the examiner presumes that the subject matter of the various claims was commonly owned as of the effective filing date of the claimed invention(s) absent any evidence to the contrary. Applicant is advised of the obligation under 37 CFR 1.56 to point out the inventor and effective filing dates of each claim that was not commonly owned as of the effective filing date of the later invention in order for the examiner to consider the applicability of 35 U.S.C. 102(b)(2)(C) for any potential 35 U.S.C. 102(a)(2) prior art against the later invention.
Claim(s) 1-6, 11-13, 21-24, 28-30, and 38 is/are rejected under 35 U.S.C. 103 as being unpatentable over Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1).
Claim 1
Regarding Claim 1, Janis discloses the following:
A system for developing a unified digital platform based virtual power banks, the system comprises [see at least Paragraph 0002 for reference to a virtual grid system for shared use of an energy producing asset by individual units of a multi-unit residential building and more specifically directed toward systems and methods for tracking and allocating shared use of an energy producing asset by individual units of a multi-unit residential building; Paragraph 0005 for reference to a virtual grid system that enables shared use of one or more distributed energy resources ("DERs") by multi-unit buildings and allocation of the benefits of such shared use to individual units; Figures 1A-1C and related text regarding example systems for shared use of a DER producing asset; Figure 2 and related text regarding an example system for shared use of a DER producing asset and communication system; Figure 3 and related text regarding an example system for monitoring and control of energy production by a shared DER producing asset; Figure 4 and related text regarding an example system for shared use of a DER producing asset; Figure 5 and related text regarding an example server system for shared use of a DER producing asset]
a memory storing program instructions [see at least Paragraph 0108 for reference to the main memory providing storage of instructions and data for programs executing on processor; Figure 22 and related text regarding item 2215 ‘Main Memory’ and item 2220 ‘Secondary Memory’]
a processor executing program instructions stored in the memory and configured to [see at least Paragraph 0106 for reference to system including one or more processors; Paragraph 0108 for reference to Main memory provides storage of instructions and data for programs executing on processor, such as one or more of the functions and/or modules discussed herein; Paragraph 0108 for reference to programs stored in the memory and executed by processor may be written and/or compiled according to any suitable language; Figure 22 and related text regarding item 2210 ‘processor’]
receive a first data type from multiple sources and store the first data type in a database as one or more record types associated with each of the multiple sources [see at least Paragraph 0043 for reference the consumption module collecting the energy usage information from the different nodes within the virtual grid system and store associated the data points to the correct physical attribute such as appliance, real estate unit, or building; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’; Examiner notes ‘energy use data’ as analogous to ‘first data’]
analyze and process the one or more record types for generating a second data type, wherein the second data type is stored in the database [see at least Paragraph 0041 for reference to the production module collecting energy production data from within the physical devices within the virtual grid system and store it in the Ivy Server ready for use by other modules; Paragraph 0054 for reference to the forecasting module using relevant data sets recorded over time inside of the ivy server to predict energy load peaks, shifts, and surplus energy generation events; Figure 15 and related text regarding ‘Primary Data Sets’ and ‘Track Energy Data Set’; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’ and item 1704 ‘Obtain DER production data during each time interval of time period’; Examiner notes ‘energy production data’ as analogous to ‘second data’]
generate virtual power banks by employing the first data type and the second data type fetched from the database, wherein the virtual power banks are power entities hosted in the unified digital platform, and the virtual power banks are associated with one or more attributes [see at least Paragraph 0050 for reference to the energy bank module is configured to communicate with the allocation module to represent the price segment of surplus energy generated on a property and how that energy fits into the load allocation outputs based on time of user energy differences; Paragraph 0064 for reference to the energy bank module is configured to keep track of surplus produced energy given real time of generation and usage comparison which will be identified in the allocation module; Paragraph 0092 for reference to the virtual bank is maintained by the server to account for variations in energy production by the one or more shared DER production assets and overall usage by the various units in the multi-unit building; Paragraph 0102 for reference to if all energy usage is met and there is still remaining DER kWh production for that time period, the remaining DER kWh are contributed to the particular time interval repository in a virtual energy bank maintained by the server; Figure 5 and related text regarding item 340 ‘energy bank module’; Figure 6 and related text regarding the load allocation module functionality with the energy bank module; Figure 7 and related text regarding the functionality of the energy bank module]
generate one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type [see at least Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
compute an optimized final weightage value of the generated virtual power banks, accessed via a the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables [see at least Paragraph 0047 for reference to the storage module identifying optimization patterns based on load balance information within the virtual grid and communicate events to the storage devices within the virtual grid system; Paragraph 0055 for reference to the weighting module weighting each unit’s usage loads against each other based on positive usage behavior and external utility pricing signals and distribute based on the determined usage; Paragraph 0072 for reference to the load forecasting module having a niche focus of optimizing energy usage behavior in the physical environment]
render a Graphical User Interface (GUI) on a user device for providing visualization functionalities related to virtual power banks for segregating power generated from renewable energy sources into a quantum based on demand on the unified digital platform [see at least Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Paragraph 0061 for reference to the interface module allowing a user with associated data inside of the virtual grid system to visualize data trends, suggestions, or preferences including suggestions on when automation events would occur, what types of events they want to be notified about, and energy usage patterns; Paragraph 0071 for reference to user interfaces to prompt energy usage behavior change based on the following: surplus onsite energy available within the physical environment as described in the load allocation module (335), utility or outside of physical environment pricing incentives to reduce energy usage such as the time of use module reference; Figure 5 and related text regarding item 295 ‘Interface API Module’, item 390 ‘Resident User Module, item 395 ‘interface module’; Examiner notes the ‘change of energy usage behavior in relation to energy availability’ as analogous to the claimed ‘segregating power generated from renewable energy sources into a quantum based on demand’]
While Janis discloses the limitations above, it does not disclose A system for developing a bi-directional cloud based unified digital platform; wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively, and wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique.
However, Sarker discloses the following:
A system for developing a bi-directional cloud based unified digital platform based virtual power banks, the system comprises [see at least Paragraph 0014 for reference to system are provided for identifying energy opportunities to optimize an objective of an entity that is a participant in the energy marketplace; Paragraph 0032 for reference to EOO system may be implemented on a variety of central computing systems such as high - performance computing systems and cloud - based computing systems]
wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to search in weight space for a set of weights that minimizes the objective function is the training process]
wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0090 for reference to the train forecaster component employing various techniques including adaptive boosting; Paragraph 0092 for reference to adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate)]
wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique [see at least Paragraph 0016 for reference to the EOO system combines the asset models along with the constraints into an overall model (e. g., for the owner of the assets) and then identifies a solution for the overall model that optimizes the energy opportunities for the assets during the target period or a series of target periods . The EOO system may apply various well - known optimization techniques to identify the solution. Some suitable techniques are described in Boyd, S. and Vandenberghe, L., “Convex Optimization,” Cambridge University Press , 2004; Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power bank system of Janis to include the optimization techniques of Sarker. Doing so would help ensure that an objective of the entity is met, as stated by Sarker (Paragraph 0014).
Claim 2
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 2, Janis discloses the following:
wherein the first data type represents data relating to power usage, requirements associated with multiple sources, data related to real-time power supply and demand received from one or more utilities and data related to multiple variables which are stored in the database, the multiple sources comprise renewable energy generators, utility retailers, end-consumers, and Electric Vehicle (EV) charging stations [see at least Paragraph 0005 for reference to the server is also configured to obtain energy usage information for each of the individual units of the multi-unit building, wherein the energy usage information for an individual unit comprises a plurality of pairs of an amount of energy used and a corresponding time interval during which the amount of energy was used; Paragraph 0036 for reference to energy usage nodes including major appliances such as an electric vehicle charger or water heater, or unitary real estate within a property such as a leased unit or separate building on a property; Paragraph 0043 for reference to the consumption module being configured to collect energy usage information from the different nodes within the virtual grid system and store the data points; Figures 1A-1C and related text regarding item 20 ‘energy producing asset’; Figure 2 and related text regarding ‘EV Charging Mngmnt Network’; Figure 5 and related text regarding item 320 EV Charging Module’; Figure 6 and related text regarding ‘Energy Bank Availability’; Figure 21 and related text regarding item 1702 ‘Obtain energy use data for all units during time period having plural intervals’]
Claim 3
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 3, Janis discloses the following:
wherein the one or more record types comprise Power Purchase Agreement (PPA) records associated with renewable energy generators, Renewable Purchase Obligation (RPO) records associated with utility retailers, Power Bank (PB) contract records associated with end-consumers, and billing and payment records associated with EV charging stations [see at least Paragraph 0078 for reference to energy usage data is collected at a residential unit level normally from a utility meter or current transformer in 15-60 minute time intervals; Paragraph 0078 for reference to public utility tariff associated to each tenant's billing profile and the Gross Utility Cost is calculated to determine the energy cost for each residential unit without the DER benefit; Figure 10 and related text regarding item 230 ‘Utility Credit Module’ including ‘Interval Energy Net Usage Amounts’]
Claim 4
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 4, Janis discloses the following:
wherein the system comprises a data analytics unit executed by the processor and configured to analyse and process the record types based on one or more functionalities to generate the second data type, the functionalities comprises enterprise resource planning, PPA and RPO record management, PB contract management, contract and transaction management, and end-consumer data management [see at least Paragraph 0003 for reference to owners of multi-unit residential buildings such as apartment buildings are increasingly installing energy producing assets such as solar panels to satisfy the demand of tenants for sustainable onsite energy production and less expensive energy expenses; Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Figure 5 and related text regarding item 390 ‘Resident User Module’]
Claim 5
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 5, Janis discloses the following:
wherein the second data type comprises data related to power usage, a quantum of power to be supplied in terms of unit capacity of the power banks, duration of supply per terms of usage of the virtual power banks, negotiated price of the virtual power banks, penalties for non- compliance related to power supply and usage per RPO, data related to end-consumer segments, and sources of power generation [see at least Paragraph 0005 for reference to the server is also configured to obtain energy usage information for each of the individual units of the multi-unit building, wherein the energy usage information for an individual unit comprises a plurality of pairs of an amount of energy used and a corresponding time interval during which the amount of energy was used; Paragraph 0036 for reference to energy usage nodes including major appliances such as an electric vehicle charger or water heater, or unitary real estate within a property such as a leased unit or separate building on a property; Paragraph 0043 for reference to the consumption module being configured to collect energy usage information from the different nodes within the virtual grid system and store the data points; Paragraph 0087 for reference to the server obtaining the kWh price of energy from the utility for each time interval during the time period; Figure 17 and related text regarding item 1622 ‘Calculate DER payment per unit using fixed price per generation credit’; Figure 21 and related text regarding item 1702 ‘Obtain energy use data for all units during time period having plural intervals’ and 1720 ‘Determine average utility price of power during time interval’]
Claim 6
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 6, Janis discloses the following:
wherein the attributes associated with the virtual power banks comprise different quantities of power (in kWh), time range for power utilization and price associated with the power based on the sources of renewable energy received from renewable energy generators [see at least Paragraph 0085 for reference to the server obtains energy use data for each unit in the multi-unit building during a time period, wherein a time period can be for example a billing cycle established by the utility; Paragraph 0087 for reference to the server obtaining kWh pricing information for the utility for each time interval during the time period; Figure 11 and related text regarding ‘Energy Cost For Time Period’; Figure 21 and related text regarding item 1702 ‘Obtain energy use data for all units during time period having plural intervals’, item 1706 ‘Obtain utility kWh rate during each time interval of time period for each unit’, and item 1708 ‘Obtain each unit energy use data during a time interval of time period’]
Claim 11
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 11, Janis discloses the following:
wherein the dynamic actionable items are generated from the first data type that includes data related to real-time power supply and demand received from one or more utilities and data related to multiple variables, and the second data type fetched from the database [see at least Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
Claim 12
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 12, Janis discloses the following:
wherein the dynamic actionable items comprise a segmented actionable item, a time-based actionable item, a peak actionable item, a penetration actionable item, a competitive actionable item, and a bulk actionable item [see at least Paragraph 0054 for reference forecasting module use relevant data sets recorded over time inside of the ivy server to predict energy load peaks, shifts, and surplus energy generation events; Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 6 and related text regarding ‘Energy Bank Availability’; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
wherein the multiple variables include season or weather, landscape conditions, current market price, frequent purchaser data, forecasted market price [see at least Paragraph 0040 for reference to data collection tasks performed by these modules are different from the energy node collection work processing tasks because they are collecting real time up to date benchmarking information on external characteristics such as utility tariff changes, regional energy market pricing trends, and external pricing indications that affect the virtual benchmarks used within the virtual grid environment; Paragraph 0058 for reference to the energy market module host a repository of external references related to energy market pricing that can be used and incorporated into the forecasting module and automation module to help user energy behavior adopt to energy market price indications; Figure 15 and related text regarding ‘Primary Data Sets’ including ‘Building Characteristics’, ‘Unit Characteristics’, ‘Weather Characteristics’, and ‘Location Characteristics’ well as the ‘Forecasted Load Variables’ factoring ‘Predicted Weather Traits’]
Claim 13
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 13, Janis discloses the following:
wherein the system comprises a virtual power bank generation unit executed by the processor and configured to generate the dynamic actionable items by employing the first data type and the second data type based on a sequence of steps comprising: determining a base value relating to a power procurement value between an end- consumer and a renewable energy generator based on a PPA between a utility and a renewable energy generator [see at least Paragraph 0063 for reference to the load allocation module applying a shared generation load across the multiple usage nodes to a virtual allocation ledger associated to the user or Multiple Unit Dwelling based on their real consumption at the time; Figure 6 and related text regarding ‘Consumption Amount MUD 1-2’]
determining a utility value associated with infrastructure usage allowance based on which a threshold value is determined, wherein the threshold value represents a minimum value below which the virtual power banks cannot be operated [see at least Paragraph 0063 for reference to the load allocation module identify all of the usage benchmarks inside of a property behind the service delivery point; Paragraph 0063 for reference to the load allocation module equally distributing the generation load across ledgers up to the usage benchmark of a node equally; Figure 6 and related text regarding the operation of the load allocation module]
determining an optimized weightage value of the virtual power banks based on the first data type and the second data type by initially identifying and processing the one or more multiple variables and determining the one or more sub-variables associated with each of the multiple variables [see at least Paragraph 0047 for reference to the storage module identifying optimization patterns based on load balance information within the virtual grid and communicate events to the storage devices within the virtual grid system; Paragraph 0063 for reference to load allocation module will also determine a property wide aggregated need for electricity from beyond the utility electricity delivery point meaning that each interval there will either be a credit generated for too much electricity produced vs consumed or a need for electricity from outside the property; Figure 6 and related text regarding the operation of the load allocation module]
Claim 21
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 21, Janis discloses the following:
wherein the system comprises a visualization unit executed by the processor and configured to render the GUI on the user device for providing the visualization functionalities comprising a dashboard, an application, and a portal for end-consumers to access the unified digital platform for operating or generating the virtual power banks [see at least Paragraph 0061 for reference to the interface module allowing a user with associated data inside of the virtual grid system to visualize data trends, suggestions, or preferences; Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Figure 5 and related text regarding item 295 ‘Interface API Module’ and item 390 ‘Resident User Module]
Claim 22
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 22, Janis discloses the following:
wherein the visualization unit provides a functionality for tracing the renewable energy source used to generate the power obtainable as virtual power banks [see at least Paragraph 0061 for reference to the interface module allowing a user with associated data inside of the virtual grid system to visualize data trends, suggestions, or preferences; Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Figure 5 and related text regarding item 295 ‘Interface API Module’ and item 390 ‘Resident User Module]
wherein the visualization unit provides a functionality of determining unused power associated with the end-consumer for re-using by listing the end-consumers having unused power [see at least Paragraph 0091 for reference to the server determines if there is excess DER kWh during the time interval; Paragraph 0092 for reference to after the server determines that there is excess DER kWh produced during the time interval, the server adds the excess DER kWh to a virtual bank for the time interval; Figure 5 and related text regarding item 295 ‘Interface API Module’ and item 390 ‘Resident User Module; Figure 21 and related text regarding item 1716 ‘Excess DER?’ and item 1718 ‘Add excess DER kWh to virtual bank for time interval’]
Claim 23
Regarding Claim 23, Janis discloses the following:
A method for developing a unified digital platform based virtual power banks, the method is implemented by a processor configured to execute instructions stored in a memory, the method comprises [see at least Paragraph 0002 for reference to the method for tracking and allocating shared use of an energy producing asset by individual units of a multi-unit residential building; Paragraph 0032 for reference to the method allowing for energy production by the energy producing asset to be tracked in time intervals and energy usage by individual units in the multi-unit building to be similarly tracked in time intervals and then usage of the energy produced by the energy producing asset to be allocated to the individual units]
receiving a first data type from multiple sources and storing the first data type in a database as one or more record types associated with each of the multiple sources [see at least Paragraph 0043 for reference the consumption module collecting the energy usage information from the different nodes within the virtual grid system and store associated the data points to the correct physical attribute such as appliance, real estate unit, or building; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’; Examiner notes ‘energy use data’ as analogous to ‘first data’]
analyzing and processing the one or more record types, wherein the record types for generating a second data type, wherein the second data types is stored in the database [see at least Paragraph 0041 for reference to the production module collecting energy production data from within the physical devices within the virtual grid system and store it in the Ivy Server ready for use by other modules; Paragraph 0054 for reference to the forecasting module using relevant data sets recorded over time inside of the ivy server to predict energy load peaks, shifts, and surplus energy generation events; Figure 15 and related text regarding ‘Primary Data Sets’ and ‘Track Energy Data Set’; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’ and item 1704 ‘Obtain DER production data during each time interval of time period’; Examiner notes ‘energy production data’ as analogous to ‘second data’]
generating virtual power banks by employing the first data type and the second data type fetched from the database, wherein the virtual power banks are power entities hosted in the unified digital platform, and wherein the virtual power banks are associated with one or more attributes [see at least Paragraph 0050 for reference to the energy bank module is configured to communicate with the allocation module to represent the price segment of surplus energy generated on a property and how that energy fits into the load allocation outputs based on time of user energy differences; Paragraph 0064 for reference to the energy bank module is configured to keep track of surplus produced energy given real time of generation and usage comparison which will be identified in the allocation module; Paragraph 0092 for reference to the virtual bank is maintained by the server to account for variations in energy production by the one or more shared DER production assets and overall usage by the various units in the multi-unit building; Paragraph 0102 for reference to if all energy usage is met and there is still remaining DER kWh production for that time period, the remaining DER kWh are contributed to the particular time interval repository in a virtual energy bank maintained by the server; Figure 5 and related text regarding item 340 ‘energy bank module’; Figure 6 and related text regarding the load allocation module functionality with the energy bank module; Figure 7 and related text regarding the functionality of the energy bank module]
generating one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type [see at least Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
computing an optimized final weightage value of the generated virtual power banks, accessed via the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub-variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables [see at least Paragraph 0047 for reference to the storage module identifying optimization patterns based on load balance information within the virtual grid and communicate events to the storage devices within the virtual grid system; Paragraph 0055 for reference to the weighting module weighting each unit’s usage loads against each other based on positive usage behavior and external utility pricing signals and distribute based on the determined usage; Paragraph 0065 for reference to rules will be equally available to all and are based on real time energy pricing weighted by utility price reference; Paragraph 0065 for reference to weighting module (365) will update based on utility time of use updates to incorporate real time market pricing into the weighting calculation; Paragraph 0067 for reference to module (330) will create an estimated energy cost figure known as “Net” that includes the savings from solar off of the “gross” cost that would have been due to the utility. This figure will be available before the utility publishes their final NET amount due. This verified NET amount will be used in the weighting module and if the weighting module used the predetermined NET amount it will calculate the difference between the two and incorporate it into the next following billing cycle; Paragraph 0072 for reference to the load forecasting module having a niche focus of optimizing energy usage behavior in the physical environment; Figure 8 and related text regarding the example functionality of a weighting module]
rendering a Graphical User Interface (GUI) on a user device for providing visualization functionalities related to virtual power banks for segregating power generated from renewable energy sources into a quantum based on demand on the unified digital platform [see at least Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Paragraph 0061 for reference to the interface module allowing a user with associated data inside of the virtual grid system to visualize data trends, suggestions, or preferences including suggestions on when automation events would occur, what types of events they want to be notified about, and energy usage patterns; Paragraph 0071 for reference to user interfaces to prompt energy usage behavior change based on the following: surplus onsite energy available within the physical environment as described in the load allocation module (335), utility or outside of physical environment pricing incentives to reduce energy usage such as the time of use module reference; Figure 5 and related text regarding item 295 ‘Interface API Module’, item 390 ‘Resident User Module, item 395 ‘interface module’; Examiner notes the ‘change of energy usage behavior in relation to energy availability’ as analogous to the claimed ‘segregating power generated from renewable energy sources into a quantum based on demand’]
While Janis discloses the limitations above, it does not disclose A method for developing a bi-directional cloud based unified digital platform; wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively, and wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique.
However, Sarker discloses the following:
A method for developing a bi-directional cloud based unified digital platform based virtual power banks [see at least Paragraph 0014 for reference to a method are provided for identifying energy opportunities to optimize an objective of an entity that is a participant in the energy marketplace; Paragraph 0032 for reference to EOO system may be implemented on a variety of central computing systems such as high - performance computing systems and cloud - based computing systems]
wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to search in weight space for a set of weights that minimizes the objective function is the training process]
wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0090 for reference to the train forecaster component employing various techniques including adaptive boosting; Paragraph 0092 for reference to adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate)]
wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique [see at least Paragraph 0016 for reference to the EOO system combines the asset models along with the constraints into an overall model (e. g., for the owner of the assets) and then identifies a solution for the overall model that optimizes the energy opportunities for the assets during the target period or a series of target periods . The EOO system may apply various well - known optimization techniques to identify the solution. Some suitable techniques are described in Boyd, S. and Vandenberghe, L., “Convex Optimization,” Cambridge University Press , 2004; Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power bank system of Janis to include the optimization techniques of Sarker. Doing so would help ensure that an objective of the entity is met, as stated by Sarker (Paragraph 0014).
Claim 24
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 24, Janis discloses the following:
wherein the record types are analyzed and processed based on one or more functionalities to generate the second data type, the functionalities comprise enterprise resource planning, Power Purchase Agreement (PPA) and Renewable Purchase Obligation (RPO) record management, Power Bank (PB) contract management, contract and transaction management, and end-consumer data management [see at least Paragraph 0078 for reference to energy usage data is collected at a residential unit level normally from a utility meter or current transformer in 15-60 minute time intervals; Paragraph 0078 for reference to public utility tariff associated to each tenant's billing profile and the Gross Utility Cost is calculated to determine the energy cost for each residential unit without the DER benefit; Figure 10 and related text regarding item 230 ‘Utility Credit Module’ including ‘Interval Energy Net Usage Amounts’]
Claim 28
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 28, Janis discloses the following:
wherein the dynamic actionable items are generated from the first data type that includes data related to real-time power supply and demand received from one or more utilities and data related to multiple variables, and the second data type fetched from the database [see at least Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
Claim 29
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 29, Janis discloses the following:
wherein the dynamic actionable items comprise a segmented actionable item, a time-based actionable item, a peak actionable item, a penetration actionable item, a competitive actionable item, and a bulk actionable item [see at least Paragraph 0054 for reference forecasting module use relevant data sets recorded over time inside of the ivy server to predict energy load peaks, shifts, and surplus energy generation events; Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 6 and related text regarding ‘Energy Bank Availability’; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
wherein the multiple variables include season or weather, landscape conditions, current market price, frequent purchaser data, forecasted market price [see at least Paragraph 0040 for reference to data collection tasks performed by these modules are different from the energy node collection work processing tasks because they are collecting real time up to date benchmarking information on external characteristics such as utility tariff changes, regional energy market pricing trends, and external pricing indications that affect the virtual benchmarks used within the virtual grid environment; Paragraph 0058 for reference to the energy market module host a repository of external references related to energy market pricing that can be used and incorporated into the forecasting module and automation module to help user energy behavior adopt to energy market price indications; Figure 15 and related text regarding ‘Primary Data Sets’ including ‘Building Characteristics’, ‘Unit Characteristics’, ‘Weather Characteristics’, and ‘Location Characteristics’ well as the ‘Forecasted Load Variables’ factoring ‘Predicted Weather Traits’]
Claim 30
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 30, Janis discloses the following:
wherein the dynamic actionable items are generated by employing the first data type and the second data type based on a sequence of steps comprising: determining a base value relating to a power procurement value between an end- consumer and a renewable energy generator based on a PPA between a utility and a renewable energy generator [see at least Paragraph 0063 for reference to the load allocation module applying a shared generation load across the multiple usage nodes to a virtual allocation ledger associated to the user or Multiple Unit Dwelling based on their real consumption at the time; Figure 6 and related text regarding ‘Consumption Amount MUD 1-2’]
determining a utility value associated with infrastructure usage allowance based on which a threshold value is determined, wherein the threshold value represents a minimum value below which the virtual power banks cannot be operated [see at least Paragraph 0063 for reference to the load allocation module identify all of the usage benchmarks inside of a property behind the service delivery point; Paragraph 0063 for reference to the load allocation module equally distributing the generation load across ledgers up to the usage benchmark of a node equally; Figure 6 and related text regarding the operation of the load allocation module]
determining an optimized weightage value of the virtual power banks based on the first data type and the second data type by initially identifying and processing the one or more multiple variables and determining the one or more sub-variables associated with each of the multiple variables [see at least Paragraph 0047 for reference to the storage module identifying optimization patterns based on load balance information within the virtual grid and communicate events to the storage devices within the virtual grid system; Paragraph 0063 for reference to load allocation module will also determine a property wide aggregated need for electricity from beyond the utility electricity delivery point meaning that each interval there will either be a credit generated for too much electricity produced vs consumed or a need for electricity from outside the property; Figure 6 and related text regarding the operation of the load allocation module]
Claim 38
Regarding Claim 38, Janis discloses the following:
A computer program product comprises: a non-transitory computer-readable medium having computer program code stored thereon, the computer-readable program code comprising instructions that, when executed by a processor, causes the processor to [see at least Paragraph 0109 for reference to the secondary medium configuration; Paragraph 0110 for reference to the secondary medium being a non-transitory computer-readable medium having computer-executable code (e.g., disclosed software modules) and/or other data stored thereon; Figure 22 and related text regarding item 2225 ‘Internal Medium’, item 2230 ‘Removable Medium’, and item 2245 ‘External Medium’]
receive a first data type from multiple sources and store the first data type in a database as one or more record types associated with each of the multiple sources [see at least Paragraph 0043 for reference the consumption module collecting the energy usage information from the different nodes within the virtual grid system and store associated the data points to the correct physical attribute such as appliance, real estate unit, or building; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’; Examiner notes ‘energy use data’ as analogous to ‘first data’]
analyze and process the one or more record types for generating a second data type, wherein the second data type is stored in the database [see at least Paragraph 0041 for reference to the production module collecting energy production data from within the physical devices within the virtual grid system and store it in the Ivy Server ready for use by other modules; Paragraph 0054 for reference to the forecasting module using relevant data sets recorded over time inside of the ivy server to predict energy load peaks, shifts, and surplus energy generation events; Figure 15 and related text regarding ‘Primary Data Sets’ and ‘Track Energy Data Set’; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’ and item 1704 ‘Obtain DER production data during each time interval of time period’; Examiner notes ‘energy production data’ as analogous to ‘second data’]
generate virtual power banks by employing the first data type and the second data type fetched from the database, wherein the virtual power banks are power entities hosted in the unified digital platform, and wherein the virtual power banks associated with one or more attributes [see at least Paragraph 0050 for reference to the energy bank module is configured to communicate with the allocation module to represent the price segment of surplus energy generated on a property and how that energy fits into the load allocation outputs based on time of user energy differences; Paragraph 0064 for reference to the energy bank module is configured to keep track of surplus produced energy given real time of generation and usage comparison which will be identified in the allocation module; Paragraph 0092 for reference to the virtual bank is maintained by the server to account for variations in energy production by the one or more shared DER production assets and overall usage by the various units in the multi-unit building; Paragraph 0102 for reference to if all energy usage is met and there is still remaining DER kWh production for that time period, the remaining DER kWh are contributed to the particular time interval repository in a virtual energy bank maintained by the server; Figure 5 and related text regarding item 340 ‘energy bank module’; Figure 6 and related text regarding the load allocation module functionality with the energy bank module; Figure 7 and related text regarding the functionality of the energy bank module]
generate one or more dynamic actionable items relating to one or more operational parameters of the virtual power banks from the first data type and the second data type [see at least Paragraph 0057 for reference to the appliance automation module communicating load balance information within the virtual grid system and push event actions to those applicable appliances such as when it is optimal to use energy based on surplus generating events in real time; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
compute an optimized final weightage value of the generated virtual power banks, accessed via the unified digital platform, by identifying and processing one or more variables that correspond to the one or more dynamic actionable items and determining one or more sub- variables associated with the variables, wherein an initial weightage value is computed for each of the sub-variables [see at least Paragraph 0047 for reference to the storage module identifying optimization patterns based on load balance information within the virtual grid and communicate events to the storage devices within the virtual grid system; Paragraph 0055 for reference to the weighting module weighting each unit’s usage loads against each other based on positive usage behavior and external utility pricing signals and distribute based on the determined usage; Paragraph 0065 for reference to rules will be equally available to all and are based on real time energy pricing weighted by utility price reference; Paragraph 0065 for reference to weighting module (365) will update based on utility time of use updates to incorporate real time market pricing into the weighting calculation; Paragraph 0067 for reference to module (330) will create an estimated energy cost figure known as “Net” that includes the savings from solar off of the “gross” cost that would have been due to the utility. This figure will be available before the utility publishes their final NET amount due. This verified NET amount will be used in the weighting module and if the weighting module used the predetermined NET amount it will calculate the difference between the two and incorporate it into the next following billing cycle; Paragraph 0072 for reference to the load forecasting module having a niche focus of optimizing energy usage behavior in the physical environment; Figure 8 and related text regarding the example functionality of a weighting module]
render a Graphical User Interface (GUI) on a user device for providing visualization functionalities related to virtual power banks for segregating power generated from renewable energy sources into a quantum based on demand on the unified digital platform [see at least Paragraph 0060 for reference to the resident user module allowing users to access their consumption data sets, virtual allocation history, avoided rate criteria and inputs, associated appliances, solar billing history, and notifications for optimizing energy behavior; Paragraph 0061 for reference to the interface module allowing a user with associated data inside of the virtual grid system to visualize data trends, suggestions, or preferences including suggestions on when automation events would occur, what types of events they want to be notified about, and energy usage patterns; Paragraph 0071 for reference to user interfaces to prompt energy usage behavior change based on the following: surplus onsite energy available within the physical environment as described in the load allocation module (335), utility or outside of physical environment pricing incentives to reduce energy usage such as the time of use module reference; Figure 5 and related text regarding item 295 ‘Interface API Module’, item 390 ‘Resident User Module, item 395 ‘interface module’; Examiner notes the ‘change of energy usage behavior in relation to energy availability’ as analogous to the claimed ‘segregating power generated from renewable energy sources into a quantum based on demand’]
While Janis discloses the limitations above, it does not disclose wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively, and wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated, and wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique.
However, Sarker discloses the following:
wherein the computed initial weightage values for each of the sub- variables is optimized by training a model iteratively [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to search in weight space for a set of weights that minimizes the objective function is the training process]
wherein results in every iteration are compared with each other by changing hyperparameters in each step until optimum results are obtained and an accurate model with less error rate is generated [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0090 for reference to the train forecaster component employing various techniques including adaptive boosting; Paragraph 0092 for reference to adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate)]
wherein the generated model is optimized by using two optimization techniques comprising a gradient descent optimization technique and a stochastic gradient descent optimization technique [see at least Paragraph 0016 for reference to the EOO system combines the asset models along with the constraints into an overall model (e. g., for the owner of the assets) and then identifies a solution for the overall model that optimizes the energy opportunities for the assets during the target period or a series of target periods . The EOO system may apply various well - known optimization techniques to identify the solution. Some suitable techniques are described in Boyd, S. and Vandenberghe, L., “Convex Optimization,” Cambridge University Press , 2004; Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power bank system of Janis to include the optimization techniques of Sarker. Doing so would help ensure that an objective of the entity is met, as stated by Sarker (Paragraph 0014).
Claim(s) 7-10 and 25-27 is/are rejected under 35 U.S.C. 103 as being unpatentable over Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1), as applied in claims 1 and 23, in view of Forbes, Jr. (U.S 2017/0358041 A1).
Claim 7
While the combination of Janis and Sarker discloses the limitations above, they do not disclose the virtual power banks employ end- to-end encryption with blockchain technology containing smart contracts for use by end- consumers, utility retailers and renewable energy generators, and wherein the smart contracts are generated based on a PPA record type of the second data type for the virtual power banks.
Regarding Claim 7, Forbes, Jr. discloses the following:
wherein the virtual power banks employ end- to-end encryption with blockchain technology containing smart contracts for use by end- consumers, utility retailers and renewable energy generators [see at least Paragraph 0250 for reference to the blockchain implementation of the smart contracts have a security via cryptography including but not limited to hashing, keys, and/or digital signatures; Paragraph 0252 for reference to smart digital contracts are self-executed between different market participants on the blockchain-based EnergyNet platform wherein, the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts)]
wherein the smart contracts are generated based on a PPA record type of the second data type for the virtual power banks [see at least Paragraph 0161 for reference to different power purchase agreements held between the EnergyNet customers, retailers, and the Clearinghouse; Paragraph 0252 for reference to smart digital contracts are self-executed between different market participants on the blockchain-based EnergyNet platform wherein, the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts); Paragraph 0254 for reference to cryptocurrency tokens are issued by the EnergyNet platform to facilitate peer-to-peer transactions between different grid elements]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include the encryption method of Forbes, Jr. Doing so simultaneously provide for a computing infrastructure, transaction platform, decentralized database, distributed account ledger, development platform, advanced energy financial settlement and marketplace, peer-to-peer network of grid elements, and a trust services layer, as stated by Forbes, Jr. (Paragraph 0250).
Claim 8
While the combination of Janis, Sarker, and Forbes, Jr. disclose the limitations above, Janis does not disclose wherein the virtual power banks are operated based on four types of smart contracts including a first type of smart contract relating to a PPA between the renewable energy generator and the utility retailer for trading renewable energy capacity, a second type of smart contract relating to a PPA between the utility retailer and the end-consumer for trading the virtual power banks, a third type of smart contract relating to a PPA between the end-consumer and an EV retailer for transfer of energy for charging of EV vehicles, and a fourth type of smart contract relating to a peer-to-peer contract between the end-consumer with other end-consumers or with retailers in selling their unused virtual power banks or self- generated renewable energy power or power stored in batteries of the EV vehicles.
Regarding Claim 8, Forbes, Jr. discloses the following:
wherein the virtual power banks are operated based on four types of smart contracts including a first type of smart contract relating to a PPA between the renewable energy generator and the utility retailer for trading renewable energy capacity, a second type of smart contract relating to a PPA between the utility retailer and the end-consumer for trading the virtual power banks, a third type of smart contract relating to a PPA between the end-consumer and an EV retailer for transfer of energy for charging of EV vehicles, and a fourth type of smart contract relating to a peer-to-peer contract between the end-consumer with other end-consumers or with retailers in selling their unused virtual power banks or self- generated renewable energy power or power stored in batteries of the EV vehicles [see at least Paragraph 0161 for reference to different power purchase agreements held between the EnergyNet customers, retailers, and the Clearinghouse; Paragraph 0163 for reference to grid elements including EV components; Paragraph 0252 for reference to smart digital contracts are self-executed between different market participants on the blockchain-based EnergyNet platform wherein, the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts); Paragraph 0254 for reference to cryptocurrency tokens are issued by the EnergyNet platform to facilitate peer-to-peer transactions between different grid elements]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include the different types of smart contracts of Forbes, Jr. Doing so simultaneously provide for a computing infrastructure, transaction platform, decentralized database, distributed account ledger, development platform, advanced energy financial settlement and marketplace, peer-to-peer network of grid elements, and a trust services layer, as stated by Forbes, Jr. (Paragraph 0250).
Claim 9
While the combination of Janis, Sarker, and Forbes, Jr. disclose the limitations above, Janis does not disclose the first type of smart contract has a unique hash key function, which is generated between the utility retailers and the renewable energy generators and provides one or more predetermined conditions comprising, quantity, price, and timeline of generated power.
Regarding Claim 9, Forbes Jr. discloses the following:
wherein the first type of smart contract has a unique hash key function, which is generated between the utility retailers and the renewable energy generators and provides one or more predetermined conditions comprising, quantity, price, and timeline of generated power [see at least Paragraph 0250 for reference to blockchain implementation of the smart contracts have a security via cryptography including but not limited to hashing, keys, and/or digital signatures wherein, hash is a unique fingerprint that is used to verify that information within the blockchain has not been altered, without the need to actually see the information itself; Paragraph 0252 for reference to the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts); Paragraph 0253 for reference to smart digital contracts in the present invention provide the transactional amount, party, and timing for payments, for which EnergyNet participants can use the built-in blockchain currency to pay for those goods; Paragraph 0256 for reference to payment methods are specified in smart contracts; Figures 108 & 109 and related text regarding the fulfilment of smart contracts]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include unique hash key function of Forbes, Jr. Doing so provide for public visibility but private inspection of the information itself, as stated by Forbes, Jr. (Paragraph 0250).
Claim 10
While the combination of Janis, Sarker, and Forbes, Jr. disclose the limitations above, Janis does not disclose the first type of smart contract is bifurcated into multiple sub-contracts which are associated with the virtual power banks and are hosted in a cloud platform, and wherein the sub-contracts have unique hash functions that ensure end-to-end encryption of the first type of contract, and wherein the hash function is backtracked for generating the first type of smart contract, each sub-contract has a transaction appending a buyer or a seller token ID.
Regarding Claim 10, Forbes, Jr. discloses the following:
wherein the first type of smart contract is bifurcated into multiple sub-contracts which are associated with the virtual power banks and are hosted in a cloud platform [see at least Paragraph 0146 for reference to the system includes a cloud-based network for distributed communication via a wireless communication antenna and processing by a plurality of mobile communication computing devices; Paragraph 0252 for reference to smart digital contracts are constructed and established within the platform by related market participants on the EnergyNet platform]
wherein the sub-contracts have unique hash functions that ensure end-to-end encryption of the first type of contract [see at least Paragraph 0250 for reference to blockchain implementation of the smart contracts have a security via cryptography including but not limited to hashing, keys, and/or digital signatures wherein, hash is a unique fingerprint that is used to verify that information within the blockchain has not been altered, without the need to actually see the information itself; Paragraph 0252 for reference to the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts)]
wherein the hash function is backtracked for generating the first type of smart contract, each sub-contract has a transaction appending a buyer or a seller token ID [see at least Paragraph 0254 for reference to the NOP tokens are based on Ethereum technology, which is an open source, blockchain-based distributed computing platform with smart contracts; Paragraph 0271 for reference to smart contracts are constructed and executed for crowdsourcing related transactions, and NOP tokens are be used in these crowdsourcing related transactions]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include sub-contract unique hash key function of Forbes, Jr. Doing so provide for public visibility but private inspection of the information itself, as stated by Forbes, Jr. (Paragraph 0250).
Claim 25
While the combination of Janis and Sarker discloses the limitations above, they do not disclose the virtual power banks are operated based on four types of smart contracts, and wherein a first type of smart contract relates to PPA between a renewable energy generator and a utility retailer for trading renewable energy capacity, a second type of smart contract relates to a PPA between the utility retailer and an end- consumer for trading the virtual power banks, a third type of smart contract relates to a PPA between the end-consumer and an EV retailer for transfer of energy for charging of EV vehicles, and a fourth type of smart contract relates to a peer-to-peer contract between consumers with other end-consumers or with retailers in selling their unused power banks or self-generated renewable energy power or power stored in batteries of the EV vehicles.
Regarding Claim 25, Forbes, Jr. discloses the following:
wherein the virtual power banks are operated based on four types of smart contracts, and wherein a first type of smart contract relates to PPA between a renewable energy generator and a utility retailer for trading renewable energy capacity, a second type of smart contract relates to a PPA between the utility retailer and an end- consumer for trading the virtual power banks, a third type of smart contract relates to a PPA between the end-consumer and an EV retailer for transfer of energy for charging of EV vehicles, and a fourth type of smart contract relates to a peer-to-peer contract between consumers with other end-consumers or with retailers in selling their unused power banks or self-generated renewable energy power or power stored in batteries of the EV vehicles [see at least Paragraph 0161 for reference to different power purchase agreements held between the EnergyNet customers, retailers, and the Clearinghouse; Paragraph 0163 for reference to grid elements including EV components; Paragraph 0252 for reference to smart digital contracts are self-executed between different market participants on the blockchain-based EnergyNet platform wherein, the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts); Paragraph 0254 for reference to cryptocurrency tokens are issued by the EnergyNet platform to facilitate peer-to-peer transactions between different grid elements]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include the different types of smart contracts of Forbes, Jr. Doing so simultaneously provide for a computing infrastructure, transaction platform, decentralized database, distributed account ledger, development platform, advanced energy financial settlement and marketplace, peer-to-peer network of grid elements, and a trust services layer, as stated by Forbes, Jr. (Paragraph 0250).
Claim 26
While the combination of Janis, Sarker, and Forbes, Jr. disclose the limitations above, Janis does not disclose the first type of smart contract has a unique hash key function, which is generated between the utility retailers and the renewable energy generators and provides one or more predetermined conditions comprising, quantity, price, and timeline of generated power.
Regarding Claim 26, Forbes, Jr. discloses the following:
wherein the first type of smart contract has a unique hash key function, which is generated between the utility retailers and the renewable energy generators and provides one or more predetermined conditions comprising, quantity, price, and timeline of generated power [see at least Paragraph 0250 for reference to blockchain implementation of the smart contracts have a security via cryptography including but not limited to hashing, keys, and/or digital signatures wherein, hash is a unique fingerprint that is used to verify that information within the blockchain has not been altered, without the need to actually see the information itself; Paragraph 0252 for reference to the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts); Paragraph 0253 for reference to smart digital contracts in the present invention provide the transactional amount, party, and timing for payments, for which EnergyNet participants can use the built-in blockchain currency to pay for those goods; Paragraph 0256 for reference to payment methods are specified in smart contracts; Figures 108 & 109 and related text regarding the fulfilment of smart contracts]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include unique hash key function of Forbes, Jr. Doing so provide for public visibility but private inspection of the information itself, as stated by Forbes, Jr. (Paragraph 0250).
Claim 27
While the combination of Janis, Sarker, and Forbes, Jr. disclose the limitations above, Janis does not disclose the first type of smart contract is bifurcated into multiple sub-contracts which are associated with the virtual power banks and are hosted in a cloud platform, and wherein the sub-contracts have unique hash functions that ensure end-to-end encryption of the first type of contract, and wherein the hash function is backtracked for generating the first type of smart contract, each sub-contract has a transaction appending a buyer or a seller token ID.
Regarding Claim 27, Forbes, Jr. discloses the following:
wherein the first type of smart contract is bifurcated into multiple sub-contracts which are associated with the virtual power banks and are hosted in a cloud platform [see at least Paragraph 0146 for reference to the system includes a cloud-based network for distributed communication via a wireless communication antenna and processing by a plurality of mobile communication computing devices; Paragraph 0252 for reference to smart digital contracts are constructed and established within the platform by related market participants on the EnergyNet platform]
wherein the sub-contracts have unique hash functions that ensure end-to-end encryption of the first type of contract [see at least Paragraph 0250 for reference to blockchain implementation of the smart contracts have a security via cryptography including but not limited to hashing, keys, and/or digital signatures wherein, hash is a unique fingerprint that is used to verify that information within the blockchain has not been altered, without the need to actually see the information itself; Paragraph 0252 for reference to the smart digital contracts in EnergyNet are similar to traditional paper-based power purchase agreements, but their terms are in a standardized form which allows them to be more easily understood and transferable to other parties (i.e., participants can buy and sell contracts)]
wherein the hash function is backtracked for generating the first type of smart contract, each sub-contract has a transaction appending a buyer or a seller token ID [see at least Paragraph 0254 for reference to the NOP tokens are based on Ethereum technology, which is an open source, blockchain-based distributed computing platform with smart contracts; Paragraph 0271 for reference to smart contracts are constructed and executed for crowdsourcing related transactions, and NOP tokens are be used in these crowdsourcing related transactions]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power banks of Janis to include sub-contract unique hash key function of Forbes, Jr. Doing so provide for public visibility but private inspection of the information itself, as stated by Forbes, Jr. (Paragraph 0250).
Claim(s) 14 and 31 is/are rejected under 35 U.S.C. 103 as being unpatentable over Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1), as applied in claim 13 and 30, in view of Willcock (US 2022/0083854 A1).
Claim 14
While the combination of Janis and Sarker discloses the limitations above, Janis does not disclose wherein the virtual power bank generation unit carries out a first optimization operation for optimizing the computed initial weightage values for each of the sub- variables, and wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation, and wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights, and wherein convergence to the optimal value provides optimal weights for one or more features associated with the model, and wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient, and wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model, and wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa; a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants, and wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate, and wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration.
However, Sarker discloses the following:
wherein the virtual power bank generation unit carries out a first optimization operation for optimizing the computed initial weightage values for each of the sub- variables [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to search in weight space for a set of weights that minimizes the objective function is the training process]
wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0089 for reference to the component loops generating training data based on historical data for a certain period for training the forecaster; Paragraph 0093 for reference to neural networks search in weight space for a set of weights that minimizes the objective function is the training process]
wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights [see at least Paragraph 0092 for reference to Adaptive boosting being an iterative process that runs multiple tests on a collection of training data; Paragraph 0092 for reference to algorithm concentrates more and more on those examples in which its predecessors tended to show mistakes. The algorithm corrects the errors made by earlier weak learners. The algorithm is adaptive because it adjusts to the error rates of its predecessors; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
wherein convergence to the optimal value provides optimal weights for one or more features associated with the model [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0090 for reference to the train forecaster component employing various techniques including adaptive boosting; Paragraph 0092 for reference to adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate)]
wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient [see at least Paragraph 0016 for reference to the EOO system combines the asset models along with the constraints into an overall model (e. g., for the owner of the assets) and then identifies a solution for the overall model that optimizes the energy opportunities for the assets during the target period or a series of target periods . The EOO system may apply various well - known optimization techniques to identify the solution. Some suitable techniques are described in Boyd, S. and Vandenberghe, L., “Convex Optimization,” Cambridge University Press , 2004; Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa [see at least Paragraph 0089 for reference to the component loops generating training data based on historical data for a certain period for training the forecaster; Paragraph 0090 for reference to machine learning techniques used to train the forecaster may be based on an autoregressive integrated moving average (“ARIMA”) model, a support vector machine ("SVM”), adaptive boosting (“AdaBoost”), a neural network model, and so on; Paragraph 0092 for reference to Adaptive boosting is an iterative process that runs multiple tests on a collection of training data. Adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate); Paragraph 0093 for reference to the neural network model search in weight space for a set of weights that minimizes the objective function is the training process. In one embodiment, the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power bank system of Janis to include the optimization techniques of Sarker. Doing so would help ensure that an objective of the entity is met, as stated by Sarker (Paragraph 0014).
While Sarker discloses the limitations above, it does not disclose wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model, and wherein the learning rate is of one or more types comprising a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants, and wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate, and wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration.
Regarding Claim 14, Willcock discloses the following:
wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation [see at least Paragraph 0042 for reference to the routine may be performed iteratively, whereby, with each iteration, the parameter continues to move closer to the optimal solution based on the training data fed to the system between iterations; Paragraph 0042 for reference to With each iteration, the set of parameters may move closer to the optimal solution based on the training data provided to the machine learning structure of the system between iterations]
wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights [see at least Paragraph 0041 for reference to the system updating the parameter w may cause the machine learning structure to output a different result for a given input , whereby a cost function of the machine learning structure is iteratively reduced or optimized over the course of several iterations of the gradient descent algorithm; Paragraph 0041 for reference to the machine learning structure may converge towards an optimal or relatively optimal set of parameters that minimizes or relatively decreases the cost function, resulting in more accurate determinations by the machine learning structure for future data inputs; Paragraph 0044 for reference to an alternative way to diminish adjustments to the machine learning structure's parameters is needed to ensure that the gradient descent algorithm converges towards a good solution]
wherein convergence to the optimal value provides optimal weights for one or more features associated with the model [see at least Paragraph 0041 for reference to the machine learning structure may converge towards an optimal or relatively optimal set of parameters that minimizes or relatively decreases the cost function, resulting in more accurate determinations by the machine learning structure for future data inputs; Paragraph 0047 for reference to the gradient descent algorithm of the present disclosure converges towards the solution faster]
wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient [see at least Paragraph 0032 for reference to the example routine for updating a parameter “w” of a gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on; Figure 2 and related text regarding the example routine for updating a parameter “w” of a gradient descent algorithm]
wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model [see at least Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Figure 1 and related text regarding item 134 ‘learning rate’]
wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa; a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants [see at least Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on; Figure 1 and related text regarding item 134 ‘learning rate’]
wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate [see at least Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on]
wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration [see at least Paragraph 0029 for reference to machine learning structure may also operate according to one or more hyper parameters, such as a number of layers, or a learning rate; Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Figure 1 and related text regarding item 134 ‘learning rate’]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Sarker to include the machine learning evaluation using sub-variables and learning rates of Willcock. Doing so parameters are improved over the course of several iterations to converge towards a value that improves an accuracy of the machine learning algorithm, as stated by Willcock (Paragraph 0002).
Claim 31
While the combination of Janis and Sarker discloses the limitations above, Janis does not disclose wherein the virtual power bank generation unit carries out a first optimization operation for optimizing the computed initial weightage values for each of the sub- variables, and wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation, and wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights, and wherein convergence to the optimal value provides optimal weights for one or more features associated with the model, and wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient, and wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model, and wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa; a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants, and wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate, and wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration.
However, Sarker discloses the following:
wherein the virtual power bank generation unit carries out a first optimization operation for optimizing the computed initial weightage values for each of the sub- variables [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to search in weight space for a set of weights that minimizes the objective function is the training process]
wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0089 for reference to the component loops generating training data based on historical data for a certain period for training the forecaster; Paragraph 0093 for reference to neural networks search in weight space for a set of weights that minimizes the objective function is the training process]
wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights [see at least Paragraph 0092 for reference to Adaptive boosting being an iterative process that runs multiple tests on a collection of training data; Paragraph 0092 for reference to algorithm concentrates more and more on those examples in which its predecessors tended to show mistakes. The algorithm corrects the errors made by earlier weak learners. The algorithm is adaptive because it adjusts to the error rates of its predecessors; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
wherein convergence to the optimal value provides optimal weights for one or more features associated with the model [see at least Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0090 for reference to the train forecaster component employing various techniques including adaptive boosting; Paragraph 0092 for reference to adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate)]
wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient [see at least Paragraph 0016 for reference to the EOO system combines the asset models along with the constraints into an overall model (e. g., for the owner of the assets) and then identifies a solution for the overall model that optimizes the energy opportunities for the assets during the target period or a series of target periods . The EOO system may apply various well - known optimization techniques to identify the solution. Some suitable techniques are described in Boyd, S. and Vandenberghe, L., “Convex Optimization,” Cambridge University Press , 2004; Paragraph 0089 for reference to the train forester component employing various machine learning techniques to train a forecaster to generate forecasts for data based on historical data to help identify energy opportunities; Paragraph 0093 for reference to the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa [see at least Paragraph 0089 for reference to the component loops generating training data based on historical data for a certain period for training the forecaster; Paragraph 0090 for reference to machine learning techniques used to train the forecaster may be based on an autoregressive integrated moving average (“ARIMA”) model, a support vector machine ("SVM”), adaptive boosting (“AdaBoost”), a neural network model, and so on; Paragraph 0092 for reference to Adaptive boosting is an iterative process that runs multiple tests on a collection of training data. Adaptive boosting transforms a weak learning algorithm (an algorithm that performs at a level only slightly better than chance) into a strong learning algorithm (an algorithm that displays a low error rate); Paragraph 0093 for reference to the neural network model search in weight space for a set of weights that minimizes the objective function is the training process. In one embodiment, the classification system may use a radial basis function (“RBF”) network and a standard gradient descent as the search technique]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the virtual power bank system of Janis to include the optimization techniques of Sarker. Doing so would help ensure that an objective of the entity is met, as stated by Sarker (Paragraph 0014).
While Sarker discloses the limitations above, it does not disclose wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model, and wherein the learning rate is of one or more types comprising a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants, and wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate, and wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration.
Regarding Claim 31, Willcock discloses the following:
wherein the first optimization operation is carried out by employing machine learning and deep learning techniques to train the model iteratively that results in a maximum and minimum function evaluation [see at least Paragraph 0042 for reference to the routine may be performed iteratively, whereby, with each iteration, the parameter continues to move closer to the optimal solution based on the training data fed to the system between iterations; Paragraph 0042 for reference to With each iteration, the set of parameters may move closer to the optimal solution based on the training data provided to the machine learning structure of the system between iterations]
wherein weights of the sub-variables are updated iteratively in opposite direction of one or more gradients of an objective function based on the gradient descent optimization technique which causes the model to find a target and converge an optimal value of the objective function based on each update of the weights [see at least Paragraph 0041 for reference to the system updating the parameter w may cause the machine learning structure to output a different result for a given input , whereby a cost function of the machine learning structure is iteratively reduced or optimized over the course of several iterations of the gradient descent algorithm; Paragraph 0041 for reference to the machine learning structure may converge towards an optimal or relatively optimal set of parameters that minimizes or relatively decreases the cost function, resulting in more accurate determinations by the machine learning structure for future data inputs; Paragraph 0044 for reference to an alternative way to diminish adjustments to the machine learning structure's parameters is needed to ensure that the gradient descent algorithm converges towards a good solution]
wherein convergence to the optimal value provides optimal weights for one or more features associated with the model [see at least Paragraph 0041 for reference to the machine learning structure may converge towards an optimal or relatively optimal set of parameters that minimizes or relatively decreases the cost function, resulting in more accurate determinations by the machine learning structure for future data inputs; Paragraph 0047 for reference to the gradient descent algorithm of the present disclosure converges towards the solution faster]
wherein gradients per iteration are updated using one sample randomly based on the stochastic gradient descent technique instead of directly computing exact value of the gradient which provides an unbiased estimate of a real gradient [see at least Paragraph 0032 for reference to the example routine for updating a parameter “w” of a gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on; Figure 2 and related text regarding the example routine for updating a parameter “w” of a gradient descent algorithm]
wherein a learning rate of the model is determined for the stochastic gradient descent optimization technique which provides flexibility to the model by discarding one or more segments of data associated with the sub-variables and a low learning rate is carried out for the model [see at least Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Figure 1 and related text regarding item 134 ‘learning rate’]
wherein the learning rate is of one or more types comprising an adaptive gradient technique (Adagrad) that provides weights with a high gradient having the low learning rate and vice versa; a Root Mean Squared Propagation (RMSprop) technique that adjusts the Adagrad technique and reduces Adagrad technique's monotonically decreasing learning rate; an Adam technique; and an Alternating Direction Method of Multipliers (ADMM) technique that provides the stochastic gradient descent variants [see at least Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on; Figure 1 and related text regarding item 134 ‘learning rate’]
wherein the gradient descent optimization technique and the AdaGrad technique varies with respect to each other based on the learning rate [see at least Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Paragraph 0049 for reference to the example routine 200 may be advantageous over alternative FTRL algorithms that involve dividing by the learning rate. It should be understood the same underlying principles may be applied to other algorithms, such as stochastic gradient descent algorithms, Adagrad, Adam, RMSProp, and so on]
wherein the learning rate is computed using all historical gradients accumulated up to a latest iteration [see at least Paragraph 0029 for reference to machine learning structure may also operate according to one or more hyper parameters, such as a number of layers, or a learning rate; Paragraph 0029 for reference to the learning rate dictating a size of the adjustments to the machine learning structure for a given iteration of the gradient descent algorithm; Paragraph 0030 for reference to the learning rate being adjusted according to the output of the gradient descent algorithm; Figure 1 and related text regarding item 134 ‘learning rate’]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Sarker to include the machine learning evaluation using sub-variables and learning rates of Willcock. Doing so parameters are improved over the course of several iterations to converge towards a value that improves an accuracy of the machine learning algorithm, as stated by Willcock (Paragraph 0002).
Claim(s) 15-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1), as applied in claim 13, in view of Bright (U.S 2018/0294649 A1).
Claim 15
While the combination of Janis and Sarker disclose the limitations above, regarding Claim 15, Janis discloses the following:
wherein the virtual power bank generation unit captures data associated with the sub-variables for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks [see at least Paragraph 0041 for reference to the production module collecting energy production data from within the physical devices within the virtual grid system and store it in the Ivy Server ready for use by other modules; Figure 15 and related text regarding ‘Primary Data Sets’ and ‘Track Energy Data Set’; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’ and item 1704 ‘Obtain DER production data during each time interval of time period’]
wherein the virtual power bank generation unit analyzes interdependency between each sub-variable with respect to another sub-variable based on the captured data [see at least Paragraph 0036 for reference to a virtual grid server and related components in a system that correlate energy data and actions within the virtual grid network according to the embodiment of the invention; Paragraph 0038 for reference to the energy data storage module correlating and compressing historical energy data]
While Janis discloses the limitations above, it does not disclose the virtual power bank generation unit analyzes interdependency between each sub-variable with respect to another sub-variable based on the captured data in a matrix form across rows and columns of the matrix.
However, Bright discloses the following:
wherein the virtual power bank generation unit analyzes interdependency between each sub-variable with respect to another sub-variable based on the captured data in a matrix form across rows and columns of the matrix [see at least Paragraph 0012 for reference to a first feature matrix representing values corresponding to the optimal electrical loads (for example the first feature matrix may represent prices corresponding to the optimal electrical loads); Paragraph 0023 for reference to a first feature matrix representing values corresponding to the load matrix , each of which optimize the first objective function while satisfying the set of constraints; Figure 2 and related text regarding item 206 ‘Identify (i) a load matrix that maximizes the first objective function and (ii) a feature matrix corresponding to the load matrix’; Figure 4 and related text regarding item 408 ‘Identify (i) a load matrix embodying a distribution of the energy resource over the power grid and (ii) a first feature matrix representing shadow values corresponding to the load matrix which optimize the objective function while satisfying the first set of constraints’]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the interdependency analysis of Janis to include the matrix analysis of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 16
While the combination of Janis, Sarker, and Bright disclose the limitations above, Janis does not disclose the virtual power bank generation unit adds the initial weightage values of the interdepending sub-variables based on the interdependency analysis and subsequently assigns a label to each of the sub-variables, and wherein the labels associated with each sub-variable are replaced with a numerical value.
Regarding Claim 16, Bright discloses the following:
wherein the virtual power bank generation unit adds the initial weightage values of the interdepending sub-variables based on the interdependency analysis and subsequently assigns a label to each of the sub-variables [see at least Paragraph 0053 for reference to the objective function may represent the cost of delivering power , the total quantity of power delivered, revenue for the generating station, revenue for the transmission substation, or any other suitable function, wherein the objective function is disclosed as a function of loads and bid values labelled by coefficients]
wherein the labels associated with each sub-variable are replaced with a numerical value [see at least Paragraph 0058 for reference to a numerical example of the objective function showing the total amount of power transmitted, wherein the results of the first optimization process in this example will be an optimal load matrix x * , and an optimal feature matrix, P*, which together optimize O]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the interdependency analysis of Janis to include the labelling of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 17
While the combination of Janis, Sarker, and Bright disclose the limitations above, Janis does not disclose wherein the virtual power bank generation unit carries out a second optimization operation for optimizing the initial weightage values by computing an average of the numerical values assigned to each of the sub-variable in either of each row or column of a matrix to generate a first weightage value for each of the variables.
Regarding Claim 17, Bright discloses the following:
wherein the virtual power bank generation unit carries out a second optimization operation for optimizing the initial weightage values by computing an average of the numerical values assigned to each of the sub-variable in either of each row or column of a matrix to generate a first weightage value for each of the variables [see at least Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the second optimization operation containing and average of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 18
While the combination of Janis, Sarker, and Bright disclose the limitations above, regarding claim 18, Janis discloses the following:
wherein the dynamic actionable items are categorized based on the identified variables [see at least Paragraph 0040 for reference to the combination of these software modules used by the virtual grid categorize these external benchmarks and the energy data nodes collected in a unique way to then process the data to creating events that drive physical value adding results within the application and enable physically applicable methodology behind the value distribution of the physically generated assets and their given value; Paragraph 0063 for reference to the load allocation module identifying all the usage benchmarks inside of a property behind the service delivery point; Paragraph 0072 for reference to load forecasting module creating prediction tracks that consist of a data formulation process that will have similar data points referenced in the figure below and is meant to associate energy usage and shift changes across energy; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
While Janis discloses the limitations above, it does not disclose the virtual power bank generation unit replaces the identified variables with one or more computed final weightage values corresponding to each of the variables, and wherein the virtual power bank generation unit carries out a third optimization operation by computing an average of the final weightage values for each of the variables associated with the dynamic actionable items for each row of a matrix.
Regarding Claim 18, Bright discloses the following:
wherein the virtual power bank generation unit replaces the identified variables with one or more computed final weightage values corresponding to each of the variables [see at least Paragraph 0006 for reference to the values of the binding functional constraints are used to calculate the replacement value (clearing value) of resources that are fully utilized (above the margin) or completely unutilized (below the margin)]
wherein the virtual power bank generation unit carries out a third optimization operation by computing an average of the final weightage values for each of the variables associated with the dynamic actionable items for each row of a matrix [see at least Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value; Paragraph 0063 for reference to some cases multiple matrices are optimal with respect to the first objective function, but only one of those optimal matrices is optimal with respect to the second objective function]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the replacement value and third optimization method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 19
While the combination of Janis, Sarker, and Bright disclose the limitations above, Janis does not disclose the virtual power bank generation unit determines a maximum average value from the computed final weightage average values associated with all the dynamic actionable items across each column of the matrix to determine a second weightage value, and wherein the second weightage value is the optimized final weightage value of the virtual power banks.
Regarding Claim 19, Bright discloses the following:
wherein the virtual power bank generation unit determines a maximum average value from the computed final weightage average values associated with all the dynamic actionable items across each column of the matrix to determine a second weightage value, and wherein the second weightage value is the optimized final weightage value of the virtual power banks [see at least Paragraph 0008 for reference to second optimization process identifies a set of value (e. g., shadow value, shadow price, or any other suitable measure of worth) that maximizes (or minimizes) the second objective function while also preserving the resource values identified by the first optimization process; Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the maximum average value method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 20
While the combination of Janis, Sarker, and Bright disclose the limitations above, Janis does not disclose the virtual power bank generation unit determines the operational parameters for the virtual power banks by carrying out a multiplication operation between the computed second weightage value and the threshold value, and then adjusting against the base value.
Regarding Claim 20, Bright discloses the following:
wherein the virtual power bank generation unit determines the operational parameters for the virtual power banks by carrying out a multiplication operation between the computed second weightage value and the threshold value, and then adjusting against the base value [see at least Paragraph 0012 for reference to receiving a first objective function and a second objective function for optimizing delivery of an energy resource over a power grid ; receiving a set of constraints representing electrical load capacities of a plurality of elements of the power grid; Paragraph 0013 for reference to the electrical load capacity is a maximum current which the electrical conduit can transmit without exceeding 100; Paragraph 0021 for reference to the second optimization is performed using at least one of a Lagrangian multiplier, an active set algorithm, an augmented Lagrangian algorithm, a conjugate gradient algorithm, and a gradient projection algorithm; Paragraph 0055 for reference to constraints include representations of maximum loads of a plurality of elements of the power grid, for example, the constraints may represent the line ratings of the transmission lines, or distribution lines of the power grid]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the operational parameter determination method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim(s) 32-37 is/are rejected under 35 U.S.C. 103 as being unpatentable over Janis (U.S 2023/0280715 A1) in view of Sarker (U.S 2019/0139159 A1) in view of Willcock (US 2022/0083854 A1), as applied in claim 31, in view of Bright (U.S 2018/0294649 A1).
Claim 32
While the combination of Janis, Sarker, and Willcock disclose the limitations above, regarding Claim 32, Janis discloses the following:
wherein data associated with the sub- variables is captured for the end-consumer who accesses the unified digital platform for obtaining the virtual power banks [see at least Paragraph 0041 for reference to the production module collecting energy production data from within the physical devices within the virtual grid system and store it in the Ivy Server ready for use by other modules; Figure 15 and related text regarding ‘Primary Data Sets’ and ‘Track Energy Data Set’; Figure 21 and related text regarding item 1202 ‘Obtain energy use data for all units during time period having plural intervals’ and item 1704 ‘Obtain DER production data during each time interval of time period’]
wherein interdependency between each sub-variable is analyzed with respect to another sub-variable based on the captured data [see at least Paragraph 0036 for reference to a virtual grid server and related components in a system that correlate energy data and actions within the virtual grid network according to the embodiment of the invention; Paragraph 0038 for reference to the energy data storage module correlating and compressing historical energy data]
While Janis discloses the limitations above, it does not disclose the interdependency between each sub-variable is analyzed with respect to another sub-variable based on the captured data in a matrix form across rows and columns of the matrix.
However, Bright discloses the following:
wherein interdependency between each sub-variable is analyzed with respect to another sub-variable based on the captured data in a matrix form across rows and columns of the matrix [see at least Paragraph 0012 for reference to a first feature matrix representing values corresponding to the optimal electrical loads (for example the first feature matrix may represent prices corresponding to the optimal electrical loads); Paragraph 0023 for reference to a first feature matrix representing values corresponding to the load matrix , each of which optimize the first objective function while satisfying the set of constraints; Figure 2 and related text regarding item 206 ‘Identify (i) a load matrix that maximizes the first objective function and (ii) a feature matrix corresponding to the load matrix’; Figure 4 and related text regarding item 408 ‘Identify (i) a load matrix embodying a distribution of the energy resource over the power grid and (ii) a first feature matrix representing shadow values corresponding to the load matrix which optimize the objective function while satisfying the first set of constraints’]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the interdependency analysis of Janis to include the matrix analysis of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 33
While the combination of Janis, Sarker, Willcock, and Bright disclose the limitations above, Janis does not disclose the initial weightage values of the interdepending sub-variables are added based on an interdependency analysis and subsequently a label is assigned to each of the sub-variables, and wherein the labels associated with each sub- variable are replaced with a numerical value.
Regarding Claim 16, Bright discloses the following:
wherein the initial weightage values of the interdepending sub-variables are added based on an interdependency analysis and subsequently a label is assigned to each of the sub-variables [see at least Paragraph 0053 for reference to the objective function may represent the cost of delivering power , the total quantity of power delivered, revenue for the generating station, revenue for the transmission substation, or any other suitable function, wherein the objective function is disclosed as a function of loads and bid values labelled by coefficients]
wherein the labels associated with each sub-variable are replaced with a numerical value [see at least Paragraph 0058 for reference to a numerical example of the objective function showing the total amount of power transmitted, wherein the results of the first optimization process in this example will be an optimal load matrix x *, and an optimal feature matrix, P*, which together optimize O]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the interdependency analysis of Janis to include the labelling of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 34
While the combination of Janis, Sarker, Willcock, and Bright disclose the limitations above, Janis does not disclose wherein a second optimization operation is carried out for optimizing the initial weightage values by computing an average of the numerical values assigned to each of the sub- variable in either of each row or column of a matrix to generate a first weightage value for each of the variables.
Regarding Claim 34, Bright discloses the following:
wherein a second optimization operation is carried out for optimizing the initial weightage values by computing an average of the numerical values assigned to each of the sub- variable in either of each row or column of a matrix to generate a first weightage value for each of the variables [see at least Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the second optimization operation containing and average of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 35
While the combination of Janis, Sarker, Willcock, and Bright disclose the limitations above, regarding claim 35, Janis discloses the following:
wherein the dynamic actionable items are categorized based on the identified variables [see at least Paragraph 0040 for reference to the combination of these software modules used by the virtual grid categorize these external benchmarks and the energy data nodes collected in a unique way to then process the data to creating events that drive physical value adding results within the application and enable physically applicable methodology behind the value distribution of the physically generated assets and their given value; Paragraph 0063 for reference to the load allocation module identifying all the usage benchmarks inside of a property behind the service delivery point; Paragraph 0072 for reference to load forecasting module creating prediction tracks that consist of a data formulation process that will have similar data points referenced in the figure below and is meant to associate energy usage and shift changes across energy; Figure 13 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’; Figure 14 and related text regarding ‘Internal Event Processing’ including dynamic actions based on the ‘Load Allocation Module’]
While Janis discloses the limitations above, it does not disclose wherein the identified variables, are replaced with the one or more computed final weightage values corresponding to each of the variables, and a third optimization operation is carried out by computing an average of the final weightage values for each of the variables associated with the dynamic actionable items for each row of the matrix.
Regarding Claim 35, Bright discloses the following:
wherein the identified variables, are replaced with the one or more computed final weightage values corresponding to each of the variables [see at least Paragraph 0006 for reference to the values of the binding functional constraints are used to calculate the replacement value (clearing value) of resources that are fully utilized (above the margin) or completely unutilized (below the margin)]
a third optimization operation is carried out by computing an average of the final weightage values for each of the variables associated with the dynamic actionable items for each row of the matrix [see at least Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value; Paragraph 0063 for reference to some cases multiple matrices are optimal with respect to the first objective function, but only one of those optimal matrices is optimal with respect to the second objective function]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the replacement value and third optimization method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 36
While the combination of Janis, Sarker, Willcock, and Bright disclose the limitations above, Janis does not disclose wherein a maximum average value is determined from the computed final weightage average values associated with all the dynamic actionable items across each column of the matrix to determine a second weightage value, and wherein the second weightage value is the optimized final weightage value of the virtual power banks
Regarding Claim 36, Bright discloses the following:
wherein a maximum average value is determined from the computed final weightage average values associated with all the dynamic actionable items across each column of the matrix to determine a second weightage value, and wherein the second weightage value is the optimized final weightage value of the virtual power banks [see at least Paragraph 0008 for reference to second optimization process identifies a set of value (e. g., shadow value, shadow price, or any other suitable measure of worth) that maximizes (or minimizes) the second objective function while also preserving the resource values identified by the first optimization process; Paragraph 0013 for reference to the second objective function is based on a sum of squares of differences between an average value and each value in the second feature matrix; Paragraph 0014 for reference to the second objective function containing the average value]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the maximum average value method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Claim 37
While the combination of Janis, Sarker, Willcock, and Bright disclose the limitations above, Janis does not disclose wherein the operational parameters for the virtual power banks are determined by carrying out a multiplication operation between the computed second weightage value and the threshold value, and then adjusting against the base value.
Regarding Claim 37, Bright discloses the following:
wherein the operational parameters for the virtual power banks are determined by carrying out a multiplication operation between the computed second weightage value and the threshold value, and then adjusting against the base value [see at least Paragraph 0012 for reference to receiving a first objective function and a second objective function for optimizing delivery of an energy resource over a power grid; receiving a set of constraints representing electrical load capacities of a plurality of elements of the power grid; Paragraph 0013 for reference to the electrical load capacity is a maximum current which the electrical conduit can transmit without exceeding 100; Paragraph 0021 for reference to the second optimization is performed using at least one of a Lagrangian multiplier, an active set algorithm, an augmented Lagrangian algorithm, a conjugate gradient algorithm, and a gradient projection algorithm; Paragraph 0055 for reference to constraints include representations of maximum loads of a plurality of elements of the power grid, for example, the constraints may represent the line ratings of the transmission lines, or distribution lines of the power grid]
Before the effective filing date, it would have been obvious to one of ordinary skill in the art to modify the optimization method of Janis to include the operational parameter determination method of Bright. Doing so would yield optimization results which are path - independent, (e.g., results which do not depend on the order in which functional constraints are processed), as stated by Bright (Paragraph 0037).
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Perera, A. T. D., et al. "Machine learning methods to assist energy system optimization." Applied energy 243 (2019): 191-205.
DOCUMENT ID
INVENTOR(S)
TITLE
US 2022/0284291 A1
Kendall, Jack David
LEARNING IN TIME VARYING, DISSIPATIVE ELECTRICAL NETWORKS
US 2021/0304306 A1
Sun et al.
Stochastic Bidding Strategy For Virtual Power Plants With Mobile Energy Storages
THIS ACTION IS MADE FINAL. Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to KRISTIN ELIZABETH GAVIN whose telephone number is (571)270-7019. The examiner can normally be reached M-F 7:30-4:30 PM EST.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jerry O'Connor can be reached at 571-272-6787. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/KRISTIN E GAVIN/Primary Examiner, Art Unit 3624
Prosecution Insights