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 .
Information Disclosure Statement
The information disclosure statement filed 25 May 2023 fails to comply with the provisions of 37 CFR 1.97, 1.98 and MPEP § 609 because NPL is missing citation information such as date, etc. It has been placed in the application file, but the information referred to therein has not been considered as to the merits. Applicant is advised that the date of any re-submission of any item of information contained in this information disclosure statement or the submission of any missing element(s) will be the date of submission for purposes of determining compliance with the requirements based on the time of filing the statement, including all certification requirements for statements under 37 CFR 1.97(e). See MPEP § 609.05(a).
Claim Rejections - 35 USC § 102
The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action:
A person shall be entitled to a patent unless –
(a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention.
Claims 1, 5-10, 12-13, and 18-20 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Lee et al. US 2021/0191348 A1.
Lee teaches:
1. A computer-implemented method of training a neural network for controlling an operation of a system having non-linear dynamics represented by partial differential equations (PDEs), wherein the neural network includes a non-linear operator of the dynamics of the system represented in a latent space by parameterized ordinary differential equations (ODEs) with parameters determined by the training, the method comprising:
collecting a digital representation of time series data indicative of instances of a function space of the system and corresponding measurements of a state of the operation of the system at different instances of time; [para. 0080, “Using the historic operational data to train the SI prediction model first may use much less computational power and time than training of the surrogate model with the DNN architecture after enough historic operational data is finally collected.”]
generating collocation points corresponding to solutions of the PDE [para. 0094, “ These variables may then be used in the identification routine to better condition the parameter estimation problem of the SI model 610.”] that represents the non-linear dynamics for a set of initial and boundary conditions on the state of the operation of the system and constraints on the operation of the system evolving from the boundary conditions according to the PDEs; [para. 0093, “The dynamic models in the zone physics sub-model of the SI model 610 may be a discretization of the indicated differential equations as follows:
PNG
media_image1.png
74
271
media_image1.png
Greyscale
with (constant) parameters M.sub.z, the thermal inertia of the zone (J/° C.), M.sub.m, the thermal inertia of the mass (J/° C.), k.sub.zm, the zone/mass heat transfer coefficient (kW/° C.), k.sub.za, the zone/ambient heat transfer coefficient (kW/° C.), and η, the ambient load scale.”]
and
training the neural network using training data including the collected time series data and the collocation points to train the parameters of the non-linear operator, [para. 0097, SI model 610 trained with both historic operational data and physics sub model]
wherein the neural network has an autoencoder architecture including
an encoder configured to encode each instance of the training data into a latent space, the non-linear operator configured to propagate the encoded instances of the training data into the latent space with transformation determined by the parameters of the non-linear operator, [para. 0119-0120 encoders] and
a decoder configured to decode the transformed encoded instances of the training data to minimize a hybrid loss function including data- driven loss between the decodings of the neural network and the collected time series data and physics-informed loss between the decodings of the neural network and solutions of the PDEs at the collocation points [para. 0119-0120; decoders].
Lee teaches:
5. The method of claim 1, wherein the non-linear operator is based on a continuous- time dynamical system. [para. 0078, “The surrogate model has a DNN architecture for continuous and adaptive learning over time using real operational data of an HVAC system that is collected from a building of interest.”]
Lee teaches:
6. The method of claim 1, further comprising fine-tuning the parameters of the non- linear operator in real-time, based on a set of expected measurements and an output of the neural network. [para. 0098, “After the model updater 620 completes re-training of the surrogate model 614, the model updater 620 may be configured to output the new system parameters predictions to the communications interface 622. As such, the BMS controller 366 may use the system parameters predictions to determine control instructions for the operation of building equipment 624 (e.g., HVAC equipment in the building of interest).”]
Lee teaches:
7. The method of claim 1, further comprising generating estimation and control commands for controlling the operation of the system. [para. 0098, “After the model updater 620 completes re-training of the surrogate model 614, the model updater 620 may be configured to output the new system parameters predictions to the communications interface 622. As such, the BMS controller 366 may use the system parameters predictions to determine control instructions for the operation of building equipment 624 (e.g., HVAC equipment in the building of interest).”]
Lee teaches:
8. The method of claim 7, wherein the generation of the estimation and control commands for controlling the operation of the system is based on a model-based control and estimation technique. [Fig. 6 surrogate model 614 used for BMS controller 366]
Lee teaches:
9. The method of claim 7, wherein the generation of the estimation and control commands for controlling the operation of the system is based on an optimization- based control and estimation technique. [para. 0055, “Demand response layer 414 can be configured to optimize resource usage (e.g., electricity use, natural gas use, water use, etc.) and/or the monetary cost of such resource usage in response to satisfy the demand of building 10. The optimization can be based on time-of-use prices, curtailment signals, energy availability, or other data received from utility providers, distributed energy generation systems 424, from energy storage 427 (e.g., hot TES 242, cold TES 244, etc.), or from other sources.”]
Lee teaches:
10. The method of claim 7, wherein the generation of the estimation and control commands for controlling the operation of the system is based on a data-driven based control and estimation technique. [Fig. 6; new operational data used to retrain model for BMS controller 366]
Lee teaches:
12. The method of claim 1, further comprising obtaining the digital representation of time series data based on use of computational fluid dynamics (CFD) simulation and experiments. [para. 0092-0094; physics based models characterizing heating/cooling zones]
Regarding system claims 13 and 18-19 as well as CRM claim 20, these apparatus claims and storage claims are rejected on the same grounds and rationale as corresponding method claims above because the apparatus claims recite the functions of steps above and the storage claims recite the storage of the method steps above.
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 2-4, 11 and 14-17 are rejected under 35 U.S.C. 103 as being unpatentable over Lee et al. US 2021/0191348 A1 in view of Dweik US 2018/0259978 A1.
Lee does not teach the following limitations, however, Dweik teaches:
2. The method of claim 1, wherein the generation of the collocation points is based on: a subset of the set of initial and boundary conditions with a structure reducing a complexity of solving the PDE; and a functional space of the system satisfying the subset of the initial and boundary conditions. [para. 0033, “In an aspect, the machine learning component 106 can also employ measured data and/or streamed data to set boundary conditions for one or more machine learning processes. For example, the machine learning component 106 can also employ measured data and/or streamed data to set boundary conditions for supply chambers and sink chambers and/or to establish driving forces for simulated physics phenomena (e.g., fluid dynamics, thermal dynamics, combustion dynamics, angular momentum, etc.).]
It would have been obvious to a person having ordinary skill in the art before the time of filing to combine the teachings of Dweik with those of Lee. A person having ordinary skill in the art would have been motivated to combine the teachings because Dweik teaches measured data can be used to set boundary conditions in simulated physic phenomenon such as fluid dynamics. (see para. 0033).
Dweik teaches:
3. The method of claim 2, wherein the structure of the subset of the initial and boundary conditions includes at least one of: sinusoidal functions, harmonic functions, periodic functions, or exponential functions. [para. 0036, “The physics behavior can also include correlations and/or behavior determined based on one or more mathematical equations associated with fluid flow such as, for example, conservation equations for mass associated with a fluid, conservation equations for momentum associated with a fluid, conservation equations for energy associated with a fluid, conservation equations for angular momentum associated with a fluid, and/or another mathematical equation associated with fluid flow.”]
Dweik teaches:
4. The method of claim 1, wherein the parameters of the non-linear operator are determined based on a probablistic approach. [para. 0034, “For example, the machine learning component 106 (e.g., one or more machine learning processes performed by the machine learning component 106) can employ a probabilistic and/or statistical-based analysis (e.g., factoring into the analysis utilities and costs) to learn and/or generate inferences with respect to the one or more 3D models generated by the modeling component 104.”]
Dweik teaches:
11. The method of claim 1, further comprising generating the parameterized ODEs based on one or more model reduction techniques, wherein the one or more model reduction techniques comprises at least one of: proper orthogonal decomposition (POD)-Galerkin projection method, or dynamic mode decomposition (DMD) method. [para. 0033; fluid dynamics]
Regarding claims 14-17, these system claims recite the functions for executing the method steps to corresponding claims above and are rejected on the same grounds and rationale as corresponding claims above.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to GARY COLLINS whose telephone number is (571)270-0473. The examiner can normally be reached Monday - Friday 1-930PM 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, Thomas Lee can be reached at (571) 272-3667. 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.
/GARY COLLINS/Examiner, Art Unit 2115