DETAILED ACTION
This action is filed in response to arguments filed on 4/29/2026.
Information Disclosure Statement
Acknowledgement is made of Applicant’s Information Disclosure Statements (IDS) form PTO-1149 filed on 5/07/2026. This IDS has been considered.
Response to Arguments
Applicant’s amendments to the title and drawings cure the prior deficiencies. Therefore the objections to the specification and drawings are withdrawn.
Applicant's arguments filed 4/29/2026 have been fully considered but they are not persuasive. Applicant argues the combined references of Krishnaswamy and Wegerich do not teach obtaining, using a model trainer, a trained plant level model; wherein the trained plant level model is obtained from the plant level model by training at least some of the equipment models in the plant level model using one or more historical datasets. Examiner notes Wegerich teaches in [0024-0025] using reference data pertaining to the equipment being monitored is input into a candidate model. Then paragraph [0032] of Wegerich teaches selecting the trained model to be deployed from those candidate models. Therefore, the trained model is obtained using historical reference data of the equipment models.
Applicant argues the dependency matrix of Krishnaswamy is not a trained plant level model. Examiner agrees. As cited in the Rejection of Claims 1, 12, and 13, the trained plant level model of Krishnaswamy is layered model 500 as described in [0058] of the prior art.
For these reasons all prior art rejections are maintained.
Claim Rejections - 35 USC § 103
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.
Claims 1-9 and 12-16 are rejected under 35 U.S.C. 103 as being unpatentable over Krishnaswamy (WO2012161945 A2) in view of Wegerich (US20070005311 A1).
Regarding Claims 1, 12, and 13, Krishnaswamy teaches a computer-implemented method for providing a model for monitoring and/or controlling an industrial plant, said industrial plant comprising a plurality of equipment (e.g. see [0006] “The computer program also includes computer readable program code for receiving data updates at the framework and notifying at least one of the applications of the data updates based on the data dependencies to support data- driven operation of the framework. The data-driven operation of the framework is configured to provide data to the applications to support performance monitoring of the equipment, analysis of the equipment's operation, and/or identification of abnormal equipment conditions”), wherein the method comprises:
- providing a plant level model of the industrial plant (e.g. see [0058] “FIGURE 5 illustrates an example layered model 500 for the large-scale framework 124 according to this disclosure. The layered model 500 illustrates how the framework 124 can be used as part of an asset manager to collect data and identify problems with rotating equipment or other equipment in an industrial facility. In this document, the term "asset" and its derivatives refer to equipment, applications, data, framework components, or other entities of an industrial facility or its control system”);
wherein the plant level model has been generated via a topology generator (e.g. see [0071] “FIGURE 8 illustrates an example logic builder 800 (i.e. a topology generator) for building models at various levels 502-510 of the large- scale framework 124 according to this disclosure. In this example, the logic builder 800 is a VISIO-based tool used to define models at any level of the layered model 500”) by selecting and interconnecting equipment models from a model library, the model library comprising computer readable equipment models for at least some of the equipment (e.g. see [0062] “In addition, the layered model 500 supports a calculation and logic execution environment. This environment could include pre- built model libraries supporting standard functions (like equipment performance calculations, instrument diagnostic parameter and limit checks, and vibration analytics modeling) . Custom model libraries can be added to support additional functionality”), and the plant level model being a topology representation of the industrial plant (e.g. see [0062] “Also, the layered model 500 supports distributed processing across multiple machines or networks and enables hierarchical modeling, with each layer 502-510 having suitable models (such as to enable fault or health monitoring on plant, unit, equipment, and sub-equipment levels to determine the health of an asset)”),
wherein the trained plant level model is usable for computing at least one performance parameter via a model executor, the at least one performance parameter being related to the industrial plant (e.g. see [0062] “In addition, the layered model 500 supports a calculation and logic execution environment. This environment could include pre- built model libraries supporting standard functions (like equipment performance calculations, instrument diagnostic parameter and limit checks, and vibration analytics modeling)”).
Krishnaswamy does not explicitly disclose obtaining, using a model trainer, a trained plant level model; wherein the trained plant level model is obtained from the plant level model by training at least some of the equipment models in the plant level model using one or more historical datasets.
In the same field of endeavor, Wegerich teaches - obtaining, using a model trainer, a trained plant level model; wherein the trained plant level model is obtained from the plant level model by training at least some of the equipment models in the plant level model (e.g. see [0014] “Model training module 140 converts the data into selected learned reference observations, which comprise the learned reference library 110”) using one or more historical datasets (e.g. see [0016] “According to this method, multivariate snapshots of sensor data are used to create a model comprising a matrix D of learned reference observations,” and [0024-0025] “turning to FIG. 2, this method is shown in a chart, wherein is plotted a sample signal as might be found in a reference data set from which in part a model is derived (along with other signals not shown). The signal 205 comprises some step function segments 210, 220 and 230, as might occur when the equipment being monitored shifts between control modes (e.g., gears or set points)… Both the original set of reference data as well as the data with perturbations is input to the candidate model, and estimates are generated. The objective is to see how badly the model estimates are influenced by the perturbed data, especially in view of how well the model makes estimates when the data is pristine”), and [0032] “Turning to FIG. 4, a method for automatically selecting a model for deployment from a set of generated candidate models is shown. In step 410, the reference data is filtered and cleaned. In step 415, a model is generated from the data. Models can vary based on tuning parameters, the type of model technology, which variables are selected to be grouped into a model, or the data snapshots used to train the model, or a combination. In step 420, the model metrics described herein are computed for the model. In step 425, if more models are to be generated, the method steps back to step 415, otherwise at step 430 the models are filtered according to their model metrics to weed out those that do not meet minimum criteria, as described below. In step 435, the remaining models are ranked according to their metrics, and a top rank model is selected for deployment in the equipment health monitoring system,” Examiner notes given the definition of ‘Snapshots’ given in [0016], these snapshots of sensor data used to train the model as taught in [0032] are historical datasets).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the trained model of Krishnaswamy with the training embodiment of Wegerich for the purpose of modeling an industrial plant with the advantage of utilizing a known metric to establish a baseline of operability that can be used to measure faults in the system.
Regarding Claim 2, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy does not explicitly disclose wherein the topology generator uses a similarity score for selecting at least one of the models.
In the same field of endeavor, Wegerich teaches wherein the topology generator uses a similarity score for selecting at least one of the models (e.g. see [0018] “Further according to a preferred embodiment of the present invention, an SBM-based model (defined as “similarity based modeling” in [0016]) can be created in real-time with each new input observation by localizing within the learned reference library 110 to those learned observations with particular relevance to the input observation, and constituting the D matrix from just those observations… A number of means of localizing may be used, including nearest neighbors to the input vector, and highest similarity scores.”
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the model selection of Krishnaswamy with the similarity scores of Wegerich for the purpose of selecting the model most aligned with the chosen industrial plant in order to obtain the most accurate monitoring data.
Regarding Claim 3, Krishnaswamy and Wegerich teach the limitations of Claim 2. Krishnaswamy does not explicitly disclose wherein the similarity score is determined based on metadata associated with the models from the model library.
In the same field of endeavor, Wegerich teaches wherein the similarity score is determined based on metadata associated with the models from the model library (e.g. see [0018] “i Further according to a preferred embodiment of the present invention, an SBM-based model can be created in real-time with each new input observation by localizing within the learned reference library 110”).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the model selection of Krishnaswamy with the similarity scores of Wegerich for the purpose of selecting the model most aligned with the chosen industrial plant in order to obtain the most accurate monitoring data.
Regarding Claim 4, Krishnaswamy and Wegerich teach the limitations of Claim 3. Krishnaswamy does not explicitly disclose wherein the metadata contains the type of equipment, the process or the reagents or products the model relates to.
In the same field of endeavor, Wegerich teaches disclose wherein the metadata contains the type of equipment, the process or the reagents or products the model relates to (e.g. see [0014] “data representative of the normal operation of equipment to be monitored, such as data from sensors on a jet engine representative of its performance throughout a flight envelope, is used in the workbench 135 to build the model. Model training module 140 converts the data into selected learned reference observations, which comprise the learned reference library 110”).
It would have been obvious to one of ordinary skill in the art before the effective filing date to combine the model library of Krishnaswamy with the metadata of Wegerich for the purpose of selecting the model most aligned with the chosen industrial plant in order to obtain the most accurate monitoring data.
Regarding Claim 5, Krishnaswamy and Wegerich teach the limitations of Claim 3. Krishnaswamy further discloses wherein the metadata is structured by an ontology (e.g. see [0072] “A model is defined in the logic definition area 806 using the functional blocks 807, which can define input and outputs of a model and calculations performed within the model. The functional blocks 807 can be selected from one or more stencils 810 (i.e. models from a model library), which identify predefined functional blocks. The functional blocks 807 could include blocks that execute particular mathematical calculations, define frequencies of interest, and reconstruct signals,” Examiner notes the functional blocks constitute metadata associated with the models that is grouped by ontology as it is groups of data combined into blocks based on the type of data and how it is used in the model).
Examiner also notes it would have been obvious to one of ordinary skill in the art to combine the model libraries of Krishnaswamy with a specific organization of the metadata as organizing data based on its type is well known in the art and Krishnaswamy teaches determining the relationships between data in the model (e.g. see [0007] “the system also includes at least one processing unit configured to identify relationships between input and output variables of multiple real-time applications associated with a framework to identify data dependencies”). Therefore it would have been obvious to one of ordinary skill in the art before the effective filling date to combine the prior art elements of model library metadata with the known method of organizing data according to ontology in order to yield the predictable result of organized data with the advantage of easily selecting a model component that is best suited for a specific industrial plant model, see MPEP 2143(I)(A).
Regarding Claim 6, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy further discloses computing, via the model executor, the at least one performance parameter using the trained plant level model (e.g. see [0066] “Here, the layer 506 can perform processing to analyze data associated with these components 702-708, such as by analyzing vibration data. The layer 508 can use the analysis results to identify faults with individual components 702-708 of the equipment 700, and the layer 510 can use the faults to identify an overall health index of the equipment 700”).
Regarding Claim 7, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy further discloses monitoring via a monitoring logic, performance of the trained plant level model (e.g. see [0074] “The use of the logic builder 800 can help to standardize maintenance practices, allow customers to insert expertise into solutions seamlessly, and integrate with manual data entry in the field. In addition, the logic builder 800 can provide support for nested models, which can be resolved recursively and evaluated at each instance level”).
Regarding Claim 8, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy does not explicitly disclose wherein the model library further comprises at least one effect model describing one or more effects related to the industrial plant, wherein an effect model refers to a model for one or more physicochemical effects or processes
In the same field of endeavor, Wegerich also teaches wherein the model library further comprises at least one effect model describing one or more effects related to the industrial plant, wherein an effect model refers to a model for one or more physicochemical effects or processes (e.g. see [0024] “turning to FIG. 2, this method is shown in a chart, wherein is plotted a sample signal as might be found in a reference data set from which in part a model is derived (along with other signals not shown). The signal 205 comprises some step function segments 210, 220 and 230, (i.e. effect models) as might occur when the equipment being monitored shifts between control modes(e.g., gears or set points)” Examiner notes the step function segments are derived from reference data used to create the model, the step functions are described as representing what “might occur” in the monitored equipment, therefore the step function segments of the reference data are effect models.)
It would have been obvious to one of ordinary skill in the art before the effective filling date to combine the model library of Krishnaswamy with the effect models of Wegerich for the purpose of monitoring industrial plant equipment with the advantage of predicting and preparing for specific outcomes for specific pieces of equipment.
Regarding Claim 9, Krishnaswamy and Wegerich teach the limitations of Claim 8. Krishnaswamy does not explicitly disclose the topology generator selecting at least one effect model from the model library.
In the same field of endeavor, Wegerich teaches disclose the topology generator selecting at least one effect model from the model library (e.g. see [0025] “Both the original set of reference data as well as the data with perturbations is input to the candidate model, and estimates are generated”).
It would have been obvious to one of ordinary skill in the art before the effective filling date to combine the model library of Krishnaswamy with the effect models of Wegerich for the purpose of monitoring industrial plant equipment with the advantage of predicting and preparing for specific outcomes for specific pieces of equipment.
Regarding Claim 14, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy further discloses a computer program, or a non-transitory computer readable medium storing the program, comprising instructions which, when the instructions are executed by any one or more suitable computing units, cause the computing units to carry out any of the steps of the method (e.g. see [0006] and [0024] “One or more servers 114 can perform various functions in the process control system 100… or example, a server 114 or operator station 116 could include at least one processing device 118a-118b, such as a processor, microprocessor, microcontroller, field programmable gate array (FPGA), or other processing or control device. A server 114 or operator station 116 could also include at least one memory 120a-120b storing instructions and data used, generated, or collected by the processing device (s),”and [0096]).
Regarding Claim 15, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy further discloses a computer storage medium, or a non-transitory computer readable medium, storing the trained plant level model as generated according to the method (e.g. see [0024] “A server 114 or operator station 116 could also include at least one memory 120a-120b storing instructions and data used, generated, or collected by the processing device (s)”).
Regarding Claim 16, Krishnaswamy and Wegerich teach the limitations of Claim 4. Krishnaswamy further discloses wherein the metadata is structured by an ontology (e.g. see [0072] “A model is defined in the logic definition area 806 using the functional blocks 807, which can define input and outputs of a model and calculations performed within the model. The functional blocks 807 can be selected from one or more stencils 810 (i.e. models from a model library), which identify predefined functional blocks. The functional blocks 807 could include blocks that execute particular mathematical calculations, define frequencies of interest, and reconstruct signals,” Examiner notes the functional blocks constitute metadata associated with the models that is grouped by ontology as it is groups of data combined into blocks based on the type of data and how it is used in the model).
Examiner also notes it would have been obvious to one of ordinary skill in the art to combine the model libraries of Krishnaswamy with a specific organization of the metadata as organizing data based on its type is well known in the art and Krishnaswamy teaches determining the relationships between data in the model (e.g. see [0007] “the system also includes at least one processing unit configured to identify relationships between input and output variables of multiple real-time applications associated with a framework to identify data dependencies”). Therefore it would have been obvious to one of ordinary skill in the art before the effective filling date to combine the prior art elements of model library metadata with the known method of organizing data according to ontology in order to yield the predictable result of organized data with the advantage of easily selecting a model component that is best suited for a specific industrial plant model, see MPEP 2143(I)(A).
Claims 10- 11 are rejected under 35 U.S.C. 103 as being unpatentable over Krishnaswamy (WO2012161945 A2) in view of Wegerich (US20070005311 A1), and in further view of Ciolfi (US10521197 B1).
Regarding Claim 10, Krishnaswamy and Wegerich teach the limitations of Claim 9. Krishnaswamy does not explicitly disclose wherein at least some of the equipment models are interconnected via one or more effect models.
In the same field of endeavor, Ciolfi teaches wherein at least some of the equipment models are interconnected via one or more effect models (e.g. see [Col 10 lines 40-44] “A sub-model also may include a group of model elements, and may be represented at a given hierarchical level by a single model reference block. A sub-model also may be saved in the library 106, and reused at other locations in the model 134 or in other models,” and [Col 13 lines 55-59] “in the case of a block that has explicitly specified its block (or port) behaviors (i.e. an effect model as it is a group of model elements describing behaviors ,i.e. effects, related to the industrial plant ), inferencing helps ensure that the attributes of the block (or port) are compatible with the attributes of the blocks (or ports) connected to it, Examiner notes this embodiment teaches connecting different model elements via the effect model of their behaviors).
It would have been obvious to one of ordinary skill in the art before the effective filling date to combine the equipment models of Krishnaswamy with the interconnection embodiment of Ciolfi for the purpose of monitoring industrial plant equipment with the advantage of monitoring a large amount of equipment at once.
Regarding Claim 11, Krishnaswamy and Wegerich teach the limitations of Claim 1. Krishnaswamy does not explicitly disclose wherein the topology generator uses one or more keywords provided via a user input for selecting at least one of the models.
In the same field of endeavor, Ciolfi teaches wherein the topology generator uses one or more keywords provided via a user input for selecting at least one of the models (e.g. see [Col 8 lines 18-25] “A user may select model elements types from the library to add instances of the selected model element types to a model being created and/or edited. The model editor 104 may perform selected operations on a model, such as open, create, edit, and save, in response to user inputs (i.e. keywords) or programmatically. The model editor 104 may include a model layout tool 115”).
It would have been obvious to one of ordinary skill in the art before the effective filling date to combine the model selection of Krishnaswamy with the keyword embodiment of Ciolfi for the purpose of monitoring industrial plant equipment with the advantage of easily selecting the model or model elements that best suit a user’s needs.
Conclusion
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 NYLA GAVIA whose telephone number is (703)756-1592. The examiner can normally be reached M-F 8:30-5:30pm.
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, Catherine Rastovski can be reached at 571-270-0349. 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.
/NYLA GAVIA/Examiner, Art Unit 2857
/Catherine T. Rastovski/Supervisory Primary Examiner, Art Unit 2857
Prosecution Insights