DETAILED ACTION
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 statements (IDS) submitted on 05/02/2023, 07/24/2023, and 02/03/2025 are in compliance with the provisions of 37 CFR 1.97. Accordingly, the information disclosure statements are being considered by the examiner.
Claim Objections
Claims 1-15 are objected to because of the following informalities:
In claim 1, line 2, “receiving at a machine learning model a training data set” should read “receiving, at a machine learning model, a training data set”
In claim 8, line 1, “the rules” should read “the one or more rules” to more properly reference “one or more rules” in line 3 of claim 4.
Dependent claims 2-15 are objected based on being directly or indirectly dependent on objected claim 1.
Appropriate correction is required.
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 5 and 18 are rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 5 recites the limitation “the value of the at least one operating parameter” in lines 4-5. There is insufficient antecedent basis for this limitation in the claim. For examination purposes, “the value of the at least one operating parameter” has been interpreted as “the value of the at least one output parameter” in reference to “a value of at least one output parameter” in line 4 of claim 4.
Claim 18 recites the limitation “the value of the at least one operating parameter” in lines 4-5. There is insufficient antecedent basis for this limitation in the claim. For examination purposes, “the value of the at least one operating parameter” has been interpreted as “the value of the at least one output parameter” in reference to “a value of at least one output parameter” in line 4 of claim 17.
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.
Claims 1-11 and 15-20 are rejected under 35 U.S.C. 103 as being unpatentable over Gadre et al. (US 2023/0195061 A1) in view of Risbeck et al. (US 2021/0041127 A1).
Regarding Claim 1,
Gadre et al. teaches a method (Fig. 4A; [0095]: "FIGS. 4A-E are flow diagrams of methods 400A-E associated with characterizing one or more components of manufacturing equipment to cause a corrective action" teaches a method for characterizing manufacturing equipment components to determine a corrective action for the manufacturing equipment) comprising:
receiving at a machine learning model a training data set describing input parameters to and corresponding output parameters from manufacturing equipment (Fig. 1; Fig. 2A; [0074]: "System 200A of FIG. 2A contains data set generator 272A (e.g., data set generator 172 of FIG. 1). Data set generator 272A creates data sets for a physics-based model (e.g., model 190 of FIG. 1). Data set generator 272A may create data sets using data retrieved from sensors associated with processing or manufacturing equipment (e.g., part quality data), data retrieved from a data store, data received from a device acting as a controller for the processing equipment, etc. In some embodiments, data set generator 272A creates training input (e.g., data input 210A) from data associated with generating processing conditions for production of substrates, e.g., processing parameter set points, part quality data associated with components of processing equipment, etc. Data set generator 272A also generates target output 220A for training a physics-based model. Target output 220A includes process sensor data 244. Target output includes sensor data collected from sensors monitoring conditions proximate to a work piece processed by manufacturing equipment. Training input data 210A and target output data 220A may be provided to a physics-based model. The physics-based model may use the training input and target output to make adjustments to parameters, coefficients, etc., to accurately predict conditions associated with a manufacturing process (e.g., predict conditions within a manufacturing chamber)" teaches a physics-based model (machine learning model) receives a training data set comprising training input data 210 to and target outputs 220 from manufacturing equipment. Fig. 1; Fig. 2A; [0078]: "In some embodiments, data set generators 272 generate data sets (e.g., training set, validating set, testing set) that includes one or more data inputs 210 (e.g., training input, validating input, testing input) and may include one or more target outputs 220 that correspond to the data inputs 210. The data set may also include mapping data that maps the data inputs 210 to the target outputs 220. Data inputs 210 may also be referred to as “features,” “attributes,” or “information.” In some embodiments, data set generators 272 may provide the data set to the training engine 182, validating engine 184, or testing engine 186 of FIG. 1, where the data set is used to train, validate, or test model 190 (e.g., a physics-based model, a machine learning model, etc.) of FIG. 1. Some embodiments of generating a training set may further be described with respect to FIG. 4A" teaches that the physics-based model is a machine learning model and that the training data set includes target outputs 220 (corresponding output parameters) that correspond to inputs 210 (input parameters)); and
training the machine learning model on the training data set using at least one learning algorithm to obtain a physics model that describes evolution of a state space of the manufacturing equipment (Fig. 1; Fig. 2A; [0074]: "System 200A of FIG. 2A contains data set generator 272A (e.g., data set generator 172 of FIG. 1). Data set generator 272A creates data sets for a physics-based model (e.g., model 190 of FIG. 1). Data set generator 272A may create data sets using data retrieved from sensors associated with processing or manufacturing equipment (e.g., part quality data), data retrieved from a data store, data received from a device acting as a controller for the processing equipment, etc. In some embodiments, data set generator 272A creates training input (e.g., data input 210A) from data associated with generating processing conditions for production of substrates, e.g., processing parameter set points, part quality data associated with components of processing equipment, etc. Data set generator 272A also generates target output 220A for training a physics-based model. Target output 220A includes process sensor data 244. Target output includes sensor data collected from sensors monitoring conditions proximate to a work piece processed by manufacturing equipment. Training input data 210A and target output data 220A may be provided to a physics-based model. The physics-based model may use the training input and target output to make adjustments to parameters, coefficients, etc., to accurately predict conditions associated with a manufacturing process (e.g., predict conditions within a manufacturing chamber)" teaches a physics-based model (machine learning model) adjust parameters/coefficients (e.g. is trained) using a training data set comprising training input data 210 (input parameters) to and target outputs 220 (output parameters) to obtain a physics model that predicts conditions (describes evolution of state space) of manufacturing equipment during a manufacturing process. Fig. 1; Fig. 2A; [0078]: "In some embodiments, data set generators 272 generate data sets (e.g., training set, validating set, testing set) that includes one or more data inputs 210 (e.g., training input, validating input, testing input) and may include one or more target outputs 220 that correspond to the data inputs 210. The data set may also include mapping data that maps the data inputs 210 to the target outputs 220. Data inputs 210 may also be referred to as “features,” “attributes,” or “information.” In some embodiments, data set generators 272 may provide the data set to the training engine 182, validating engine 184, or testing engine 186 of FIG. 1, where the data set is used to train, validate, or test model 190 (e.g., a physics-based model, a machine learning model, etc.) of FIG. 1. Some embodiments of generating a training set may further be described with respect to FIG. 4A" teaches that the physics-based model is a machine learning model that is trained using the training data set that includes target outputs 220 that correspond to inputs 210 (i.e. trained using at least one learning algorithm since training with inputs and corresponding target outputs is a supervised machine learning algorithm)).
Gadre et al. does not appear to explicitly teach configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters; and training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges.
However, Risbeck et al. teaches configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters (Fig. 6; [0113]-[0114]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building … Still referring to FIG. 6, controller 204 is shown to include a model trainer 640. Model trainer 640 can be configured to perform a training procedure to train controller model 414, physics model 412, and disturbance model 402. In some embodiments, model trainer 640 performs one training procedure to train controller model 414 and another training procedure to train both physics model 412 and disturbance model 402. The training procedure for controller model 414 may include fitting controller model 414 to a set of training data (e.g., inputs u and outputs y of controller model 414) to generate values of the trainable parameters of controller model 414. The trainable parameters of controller model 414 may include, for example, the α and β parameters of Eq. 13, the parameters of the function c(⋅) of Eq. 14, or any other trainable parameters that may exist in controller model 414. After controller model 414 is fully trained, controller model 414 can be used to translate a zone temperature Tz and/or a heating/cooling duty
Q
˙
HVAC into a temperature setpoint Tsp required to achieve the zone temperature Tz and/or the heating/cooling duty
Q
˙
HVAC" teaches configuring a controller model 414 of the controller (machine learning-based controller agent) that can be used to control manufacturing processes along with a physics model. Fig. 6; [0124]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 of the controller (machine learning-based controller agent) uses (generates commands for) the physics model to generate and impose constraints (modify settings) on the model predictive control process (e.g. simulation of manufacturing equipment) such that the physics model predicts the output evolution of the states (predicted output parameters) in response to the provided inputs); and
training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges (Fig. 6; [0113]-[0114]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building … Still referring to FIG. 6, controller 204 is shown to include a model trainer 640. Model trainer 640 can be configured to perform a training procedure to train controller model 414, physics model 412, and disturbance model 402. In some embodiments, model trainer 640 performs one training procedure to train controller model 414 and another training procedure to train both physics model 412 and disturbance model 402. The training procedure for controller model 414 may include fitting controller model 414 to a set of training data (e.g., inputs u and outputs y of controller model 414) to generate values of the trainable parameters of controller model 414. The trainable parameters of controller model 414 may include, for example, the α and β parameters of Eq. 13, the parameters of the function c(⋅) of Eq. 14, or any other trainable parameters that may exist in controller model 414. After controller model 414 is fully trained, controller model 414 can be used to translate a zone temperature Tz and/or a heating/cooling duty
Q
˙
HVAC into a temperature setpoint Tsp required to achieve the zone temperature Tz and/or the heating/cooling duty
Q
˙
HVAC" teaches that the controller model 414 of the controller (machine learning-based controller agent) is trained (e.g. using at least one other learning algorithm) using the inputs u and corresponding outputs y (settings and corresponding predicted output parameters) such that the controller model 414 in response to inputs maintains a value of a predicted temperature setpoint range (predicted output parameters within respective predefined ranges)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters; and training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 2,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Risbeck et al. further teaches further comprising configuring the machine-learning-based controller agent to generate commands for the manufacturing equipment responsive to values of predicted output parameters from the physics model such that the manufacturing equipment executes the commands (Fig. 4; Fig. 6; [0060]: "Monolithic system model 410 is shown to include a physics model 412 and a controller model 414. Physics model 412 may model the temperature dynamics of building zone 202. For example, physics model 412 may receive the heat disturbance
Q
˙
other as well as the heating or cooling duty
Q
˙
HVAC for HVAC equipment 206 as inputs and may predict the temperature Tz of building zone 202 as an output. Controller model 414 may model the behavior of a closed loop feedback controller that generates a control signal for HVAC equipment 206 based on the temperature setpoint Tsp and the zone temperature Tz. For example, controller model 414 may receive the temperature setpoint Tsp and the zone temperature Tz as inputs and may output the heating or cooling duty
Q
˙
HVAC for HVAC equipment 206 as an output" teaches that the controller model 414 of the controller (machine learning-based controller agent) generates control signals (commands) for the HVAC equipment (manufacturing equipment) to execute in response to predicted temperature outputs (predicted output parameters) from the physics model. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising configuring the machine-learning-based controller agent to generate commands for the manufacturing equipment responsive to values of predicted output parameters from the physics model such that the manufacturing equipment executes the commands as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 3,
Gadre et al. in view of Risbeck et al. teaches the method of claim 2.
In addition, Risbeck et al. further teaches further comprising configuring the machine-learning-based controller agent to generate the commands for the manufacturing equipment responsive to input parameters to and corresponding output parameters from the manufacturing equipment (Fig. 2; Fig. 6; [0050]: "Controller 204 receives the temperature measurements Tz and Ta from sensors 208-210 and provides control signals to HVAC equipment 206. In some embodiments, the control signals include heating or cooling duties for HVAC equipment 206. Advantageously, controller 204 may consider the sources of heat transfer provided by heat load 204 (i.e.,
Q
˙
other), building mass 212 (i.e.,
Q
˙
m), and ambient air 218 (i.e.,
Q
˙
a) on the zone air temperature Tz and may operate HVAC equipment 206 to provide a suitable amount of heating or cooling HVAC to maintain the zone air temperature Tz within an acceptable range" teaches that the controller (machine learning-based controller agent) generates control signals (commands) for the HVAC equipment (manufacturing equipment) to execute in response to input parameters (e.g. heat load, building mass, ambient air) and corresponding predicted temperature measurements (corresponding output parameters) from HVAC equipment and sensors (manufacturing equipment). [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising configuring the machine-learning-based controller agent to generate the commands for the manufacturing equipment responsive to input parameters to and corresponding output parameters from the manufacturing equipment as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 4,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Risbeck et al. further teaches further comprising: wherein the configuring includes receiving at the machine-learning-based controller agent one or more rules defining control actions for the manufacturing equipment to be taken responsive to a value of at least one output parameter from the manufacturing equipment being outside a predefined range (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) uses the physics model to generate and impose constraints (rules defining control actions) on the model predictive control process such that the physics model predicts output evolution of the states (predicted output parameters) in response to the provided inputs for the HVAC equipment (manufacturing equipment) respect the predefined range. Fig. 6; [0137]: "Model predictive controller 622 can execute the model predictive control process (i.e., solving the linear optimization problem subject to the constraints) to determine the optimal values of the decision variables, including the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t, for each time step t in the optimization period. Model predictive controller 622 can use physics model 412 and/or the constraint in Eq. 41 to predict the system states Tzt and Tm,t resulting from the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t over the optimization period" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) optimizes (executes control actions) the model predictive control process to ensure the predicted system states generated from the physics model and constraints (rules) respect the predefined range. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: wherein the configuring includes receiving at the machine-learning-based controller agent one or more rules defining control actions for the manufacturing equipment to be taken responsive to a value of at least one output parameter from the manufacturing equipment being outside a predefined range as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 5,
Gadre et al. in view of Risbeck et al. teaches the method of claim 4.
In addition, Risbeck et al. further teaches further comprising: receiving machine-learning-based controller agent time series data describing operating states of the manufacturing equipment (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) receives inputs at a plurality of time steps over a time period (time series data) describing the states of the temperature of the building zone to be used by the physics model to predict the output evolution of the states for the HVAC equipment (manufacturing equipment). [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)); and
responsive to the operating states indicating the value of the at least one operating parameter is outside the predefined range, generating by the machine-learning-based controller agent a command for the manufacturing equipment to execute at least one of the control actions such that the manufacturing equipment performs the at least one of the control actions (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) uses the physics model to generate and impose constraints (rules defining control actions) on the model predictive control process such that the physics model predicts output evolution of the states (predicted output parameters) in response to the provided inputs for the HVAC equipment (manufacturing equipment) respect the predefined range. Fig. 6; [0137]: "Model predictive controller 622 can execute the model predictive control process (i.e., solving the linear optimization problem subject to the constraints) to determine the optimal values of the decision variables, including the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t, for each time step t in the optimization period. Model predictive controller 622 can use physics model 412 and/or the constraint in Eq. 41 to predict the system states Tzt and Tm,t resulting from the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t over the optimization period" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) optimizes (executes control actions) the model predictive control process to ensure the predicted system states generated from the physics model and constraints (rules) respect the predefined range. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: receiving machine-learning-based controller agent time series data describing operating states of the manufacturing equipment; and responsive to the operating states indicating the value of the at least one operating parameter is outside the predefined range, generating by the machine-learning-based controller agent a command for the manufacturing equipment to execute at least one of the control actions such that the manufacturing equipment performs the at least one of the control actions as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 6,
Gadre et al. in view of Risbeck et al. teaches the method of claim 4.
In addition, Risbeck et al. further teaches wherein the one or more rules are obtained from a rules controller (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 (rules controller) generates and imposes a set of constraints (rules)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the one or more rules are obtained from a rules controller as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 7,
Gadre et al. in view of Risbeck et al. teaches the method of claim 4.
In addition, Risbeck et al. further teaches further comprising receiving machine-learning-based controller agent input modifying the one or more rules in real time (Fig. 6; Fig. 7; [0169]: "Still referring to FIG. 7, process 700 is shown to include predicting the zone temperature Tz over the time period by performing model predictive control using the linear physics model and the predicted values of the heat load disturbance
Q
˙
other over the time period (step 708). In some embodiments, step 708 is performed by model predictive controller 622 as described with reference to FIG. 6. The model predictive control process performed in step 708 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Step 708 may include optimizing the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 of the controller (machine learning-based controller agent) performs the model predictive control process over a time period. Fig. 6; Eq. 41; [0126]: "Model predictive controller 622 may optimize the objective function J subject to the following constraint:
PNG
media_image1.png
42
592
media_image1.png
Greyscale
which may be based on physics model 412. This constraint defines the system states xt+1 at the next time step t+1 as a function ƒ(⋅) of the system states xt, the inputs xt, and the exogenous parameters pt at the current time step t" teaches that the constraint (rule) is modified in real time).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising receiving machine-learning-based controller agent input modifying the one or more rules in real time as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 8,
Gadre et al. in view of Risbeck et al. teaches the method of claim 4.
In addition, Risbeck et al. further teaches wherein the settings incorporate the rules (Fig. 6; [0124]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 of the controller (machine learning-based controller agent) uses (generates commands for) the physics model to generate and impose constraints (modify settings to incorporate the constraints/rules) on the model predictive control process (e.g. simulation of manufacturing equipment) such that the physics model predicts the output evolution of the states (predicted output parameters) in response to the provided inputs).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the settings incorporate the rules as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 9,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Risbeck et al. further teaches further comprising operating the manufacturing equipment with a rules controller to generate the training data set (Fig. 6; [0124]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 (rules controller) uses (generates commands for) the physics model to generate and impose constraints (rules) on the model predictive control process such that the physics model predicts the output dynamic temperature response (e.g. temperature Tz is predicted) in response to the provided inputs for the HVAC equipment (manufacturing equipment). Fig. 6; [0098]: "a set of training data that includes values of the ambient temperature Ta, the weather forecast (e.g., cloudiness C), the time and date t, the electric load of building zone 202, measurements of the zone temperature Tz, and values of the heating or cooling duty
Q
˙
HVAC. All of these variables can be readily measured, observed, or predicted and may be available for use as training data" teaches that the predicted measurement of the temperature Tz may be used in the training data set. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for operating the manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising operating the manufacturing equipment with a rules controller to generate the training data set as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 10,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Gadre et al. further teaches wherein the input parameters include active control parameters, endogenous parameters, and exogenous parameters of the manufacturing equipment (Fig. 4A; [0099]-[0100]: "At block 402, processing logic generates first data input (e.g., first training input, first validating input) that may include sensor data, metrology data (e.g., film properties such as thickness, material composition, optical properties, roughness, and so on), processing parameter data, part quality data, etc. In some embodiments, the first data input may include a first set of features for types of data and a second data input may include a second set of features for types of data (e.g., as described with respect to FIG. 3). … For example, method 400A may be used to generate data sets for a machine learning model configured to accept as input target processing parameters (e.g., processing set points) and measured performance data (e.g., sensor data, metrology data, etc.) and produce as output predictions of values of one or more quality parameters of one or more components of the manufacturing equipment. At block 402, training input comprising sensor data, combinations of sensor data, features of sensor data, etc., may be generated" teaches that the input data (input parameters) is generated based on sensor data ,metrology data, processing parameter data, and historical manufacturing parameters. Fig. 1; [0028]: "Sensor data 142 may include historical sensor data and current sensor data. Manufacturing equipment 124 may be configured according to manufacturing parameters 150. Manufacturing parameters 150 may be associated with or indicative of parameters such as hardware parameters (e.g., settings or components (e.g., size, type, etc.) of the manufacturing equipment 124) and/or process parameters of the manufacturing equipment. Manufacturing parameters 150 may include historical manufacturing data and/or current manufacturing data. Manufacturing parameters 150 may be indicative of input settings to the manufacturing device (e.g., heater power, gas flow, etc.). Sensor data 142 and/or manufacturing parameters 150 may be provided while the manufacturing equipment 124 is performing manufacturing processes (e.g., equipment readings when processing products)" teaches that the manufacturing equipment is configured according to manufacturing parameters that include active control parameters, endogenous parameters, and exogenous parameters of the manufacturing equipment.).
Regarding Claim 11,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Gadre et al. further teaches wherein the output parameters include feature parameters of components produced by the manufacturing equipment (Fig. 4A; [0101]: "At block 403, processing logic generates a first target output for one or more of the data inputs (e.g., first data input). In some embodiments, the first target output is part quality data. In some embodiments, the first target output is indicative of manufacturing equipment performance. In some embodiments, the first target output is data indicative of a corrective action" teaches that the target output (output parameter) includes data about part quality or manufacturing equipment performance for the manufacturing equipment (e.g. target outputs include feature parameters of components produced by the manufacturing equipment)).
Regarding Claim 15,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
In addition, Gadre et al. further teaches wherein the at least one learning algorithm is a supervised learning algorithm (Fig. 1; Fig. 2A; [0074]: "System 200A of FIG. 2A contains data set generator 272A (e.g., data set generator 172 of FIG. 1). Data set generator 272A creates data sets for a physics-based model (e.g., model 190 of FIG. 1). Data set generator 272A may create data sets using data retrieved from sensors associated with processing or manufacturing equipment (e.g., part quality data), data retrieved from a data store, data received from a device acting as a controller for the processing equipment, etc. In some embodiments, data set generator 272A creates training input (e.g., data input 210A) from data associated with generating processing conditions for production of substrates, e.g., processing parameter set points, part quality data associated with components of processing equipment, etc. Data set generator 272A also generates target output 220A for training a physics-based model. Target output 220A includes process sensor data 244. Target output includes sensor data collected from sensors monitoring conditions proximate to a work piece processed by manufacturing equipment. Training input data 210A and target output data 220A may be provided to a physics-based model. The physics-based model may use the training input and target output to make adjustments to parameters, coefficients, etc., to accurately predict conditions associated with a manufacturing process (e.g., predict conditions within a manufacturing chamber)" teaches a physics-based model (machine learning model) adjust parameters/coefficients (e.g. is trained) using a training data set comprising training input data 210 (input parameters) to and target outputs 220 (output parameters) to obtain a physics model that predicts conditions (describes evolution of state space) of manufacturing equipment during a manufacturing process. Fig. 1; Fig. 2A; [0078]: "In some embodiments, data set generators 272 generate data sets (e.g., training set, validating set, testing set) that includes one or more data inputs 210 (e.g., training input, validating input, testing input) and may include one or more target outputs 220 that correspond to the data inputs 210. The data set may also include mapping data that maps the data inputs 210 to the target outputs 220. Data inputs 210 may also be referred to as “features,” “attributes,” or “information.” In some embodiments, data set generators 272 may provide the data set to the training engine 182, validating engine 184, or testing engine 186 of FIG. 1, where the data set is used to train, validate, or test model 190 (e.g., a physics-based model, a machine learning model, etc.) of FIG. 1. Some embodiments of generating a training set may further be described with respect to FIG. 4A" teaches that the physics-based model is a machine learning model that is trained using the training data set that includes target outputs 220 that correspond to inputs 210 (i.e. trained using at least one learning algorithm since training with inputs and corresponding target outputs is a supervised machine learning algorithm)).
Regarding Claim 16,
Gadre et al. teaches a method (Fig. 4A; [0095]: "FIGS. 4A-E are flow diagrams of methods 400A-E associated with characterizing one or more components of manufacturing equipment to cause a corrective action" teaches a method for characterizing manufacturing equipment components to determine a corrective action for the manufacturing equipment) comprising:
training a machine learning model on a training data set, that describes input parameters to and corresponding output parameters from manufacturing equipment, using at least one learning algorithm to obtain a physics model that describes evolution of a state space of the manufacturing equipment (Fig. 1; Fig. 2A; [0074]: "System 200A of FIG. 2A contains data set generator 272A (e.g., data set generator 172 of FIG. 1). Data set generator 272A creates data sets for a physics-based model (e.g., model 190 of FIG. 1). Data set generator 272A may create data sets using data retrieved from sensors associated with processing or manufacturing equipment (e.g., part quality data), data retrieved from a data store, data received from a device acting as a controller for the processing equipment, etc. In some embodiments, data set generator 272A creates training input (e.g., data input 210A) from data associated with generating processing conditions for production of substrates, e.g., processing parameter set points, part quality data associated with components of processing equipment, etc. Data set generator 272A also generates target output 220A for training a physics-based model. Target output 220A includes process sensor data 244. Target output includes sensor data collected from sensors monitoring conditions proximate to a work piece processed by manufacturing equipment. Training input data 210A and target output data 220A may be provided to a physics-based model. The physics-based model may use the training input and target output to make adjustments to parameters, coefficients, etc., to accurately predict conditions associated with a manufacturing process (e.g., predict conditions within a manufacturing chamber)" teaches a physics-based model (machine learning model) adjust parameters/coefficients (e.g. is trained) using a training data set comprising training input data 210 (input parameters) to and target outputs 220 (output parameters) from manufacturing equipment to obtain a physics model that predicts conditions (describes evolution of state space) of manufacturing equipment during a manufacturing process. Fig. 1; Fig. 2A; [0078]: "In some embodiments, data set generators 272 generate data sets (e.g., training set, validating set, testing set) that includes one or more data inputs 210 (e.g., training input, validating input, testing input) and may include one or more target outputs 220 that correspond to the data inputs 210. The data set may also include mapping data that maps the data inputs 210 to the target outputs 220. Data inputs 210 may also be referred to as “features,” “attributes,” or “information.” In some embodiments, data set generators 272 may provide the data set to the training engine 182, validating engine 184, or testing engine 186 of FIG. 1, where the data set is used to train, validate, or test model 190 (e.g., a physics-based model, a machine learning model, etc.) of FIG. 1. Some embodiments of generating a training set may further be described with respect to FIG. 4A" teaches that the physics-based model is a machine learning model that is trained using the training data set that includes target outputs 220 (corresponding output parameters) that correspond to inputs 210 (input parameters) (i.e. trained using at least one learning algorithm since training with inputs and corresponding target outputs is a supervised machine learning algorithm)).
Gadre et al. does not appear to explicitly teach configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters; training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges; and configuring the machine-learning-based controller agent to generate commands for the manufacturing equipment responsive to values of predicted output parameters from the physics model such that the manufacturing equipment executes the commands.
However, Risbeck et al. teaches configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters (Fig. 6; [0113]-[0114]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building … Still referring to FIG. 6, controller 204 is shown to include a model trainer 640. Model trainer 640 can be configured to perform a training procedure to train controller model 414, physics model 412, and disturbance model 402. In some embodiments, model trainer 640 performs one training procedure to train controller model 414 and another training procedure to train both physics model 412 and disturbance model 402. The training procedure for controller model 414 may include fitting controller model 414 to a set of training data (e.g., inputs u and outputs y of controller model 414) to generate values of the trainable parameters of controller model 414. The trainable parameters of controller model 414 may include, for example, the α and β parameters of Eq. 13, the parameters of the function c(⋅) of Eq. 14, or any other trainable parameters that may exist in controller model 414. After controller model 414 is fully trained, controller model 414 can be used to translate a zone temperature Tz and/or a heating/cooling duty
Q
˙
HVAC into a temperature setpoint Tsp required to achieve the zone temperature Tz and/or the heating/cooling duty
Q
˙
HVAC" teaches configuring a controller model 414 of the controller (machine learning-based controller agent) that can be used to control manufacturing processes along with a physics model. Fig. 6; [0124]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 of the controller (machine learning-based controller agent) uses (generates commands for) the physics model to generate and impose constraints (modify settings) on the model predictive control process (e.g. simulation of manufacturing equipment) such that the physics model predicts the output evolution of the states (predicted output parameters) in response to the provided inputs);
training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges (Fig. 6; [0113]-[0114]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building … Still referring to FIG. 6, controller 204 is shown to include a model trainer 640. Model trainer 640 can be configured to perform a training procedure to train controller model 414, physics model 412, and disturbance model 402. In some embodiments, model trainer 640 performs one training procedure to train controller model 414 and another training procedure to train both physics model 412 and disturbance model 402. The training procedure for controller model 414 may include fitting controller model 414 to a set of training data (e.g., inputs u and outputs y of controller model 414) to generate values of the trainable parameters of controller model 414. The trainable parameters of controller model 414 may include, for example, the α and β parameters of Eq. 13, the parameters of the function c(⋅) of Eq. 14, or any other trainable parameters that may exist in controller model 414. After controller model 414 is fully trained, controller model 414 can be used to translate a zone temperature Tz and/or a heating/cooling duty
Q
˙
HVAC into a temperature setpoint Tsp required to achieve the zone temperature Tz and/or the heating/cooling duty
Q
˙
HVAC" teaches that the controller model 414 of the controller (machine learning-based controller agent) is trained (e.g. using at least one other learning algorithm) using the inputs u and corresponding outputs y (settings and corresponding predicted output parameters) such that the controller model 414 in response to inputs maintains a value of a predicted temperature setpoint range (predicted output parameters within respective predefined ranges)); and
configuring the machine-learning-based controller agent to generate commands for the manufacturing equipment responsive to values of predicted output parameters from the physics model such that the manufacturing equipment executes the commands (Fig. 4; Fig. 6; [0060]: "Monolithic system model 410 is shown to include a physics model 412 and a controller model 414. Physics model 412 may model the temperature dynamics of building zone 202. For example, physics model 412 may receive the heat disturbance
Q
˙
other as well as the heating or cooling duty
Q
˙
HVAC for HVAC equipment 206 as inputs and may predict the temperature Tz of building zone 202 as an output. Controller model 414 may model the behavior of a closed loop feedback controller that generates a control signal for HVAC equipment 206 based on the temperature setpoint Tsp and the zone temperature Tz. For example, controller model 414 may receive the temperature setpoint Tsp and the zone temperature Tz as inputs and may output the heating or cooling duty
Q
˙
HVAC for HVAC equipment 206 as an output" teaches that the controller model 414 of the controller (machine learning-based controller agent) generates control signals (commands) for the HVAC equipment (manufacturing equipment) to execute in response to predicted temperature outputs (predicted output parameters) from the physics model. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate configuring a machine-learning-based controller agent to generate commands for the physics model that modify settings of a simulation of the manufacturing equipment by the physics model such that, responsive to input data, the physics model generates corresponding predicted output parameters; training the machine-learning-based controller agent on the settings and corresponding predicted output parameters using at least one other learning algorithm such that, responsive to input data, the machine-learning-based controller agent maintains values of the predicted output parameters within respective predefined ranges; and configuring the machine-learning-based controller agent to generate commands for the manufacturing equipment responsive to values of predicted output parameters from the physics model such that the manufacturing equipment executes the commands as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 17,
Gadre et al. in view of Risbeck et al. teaches the method of claim 16.
In addition, Risbeck et al. further teaches further comprising: wherein the configuring includes receiving at the machine-learning-based controller agent one or more rules defining control actions for the manufacturing equipment to be taken responsive to a value of at least one output parameter from the manufacturing equipment being outside a predefined range (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) uses the physics model to generate and impose constraints (rules defining control actions) on the model predictive control process such that the physics model predicts output evolution of the states (predicted output parameters) in response to the provided inputs for the HVAC equipment (manufacturing equipment) respect the predefined range. Fig. 6; [0137]: "Model predictive controller 622 can execute the model predictive control process (i.e., solving the linear optimization problem subject to the constraints) to determine the optimal values of the decision variables, including the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t, for each time step t in the optimization period. Model predictive controller 622 can use physics model 412 and/or the constraint in Eq. 41 to predict the system states Tzt and Tm,t resulting from the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t over the optimization period" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) optimizes (executes control actions) the model predictive control process to ensure the predicted system states generated from the physics model and constraints (rules) respect the predefined range. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: wherein the configuring includes receiving at the machine-learning-based controller agent one or more rules defining control actions for the manufacturing equipment to be taken responsive to a value of at least one output parameter from the manufacturing equipment being outside a predefined range as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 18,
Gadre et al. in view of Risbeck et al. teaches the method of claim 17.
In addition, Risbeck et al. further teaches further comprising: receiving machine-learning-based controller agent time series data describing operating states of the manufacturing equipment (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) receives inputs at a plurality of time steps over a time period (time series data) describing the states of the temperature of the building zone to be used by the physics model to predict the output evolution of the states for the HVAC equipment (manufacturing equipment). [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)); and
responsive to the operating states indicating the value of the at least one operating parameter is outside the predefined range, generating by the machine-learning-based controller agent a command for the manufacturing equipment to execute at least one of the control actions such that the manufacturing equipment performs the at least one of the control actions (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) uses the physics model to generate and impose constraints (rules defining control actions) on the model predictive control process such that the physics model predicts output evolution of the states (predicted output parameters) in response to the provided inputs for the HVAC equipment (manufacturing equipment) respect the predefined range. Fig. 6; [0137]: "Model predictive controller 622 can execute the model predictive control process (i.e., solving the linear optimization problem subject to the constraints) to determine the optimal values of the decision variables, including the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t, for each time step t in the optimization period. Model predictive controller 622 can use physics model 412 and/or the constraint in Eq. 41 to predict the system states Tzt and Tm,t resulting from the optimal values of Qt, Qh,t, Qc,t, and/or
Q
˙
HVAC,t over the optimization period" teaches that the model predictive controller 622 of the controller (machine learning-based controller agent) optimizes (executes control actions) the model predictive control process to ensure the predicted system states generated from the physics model and constraints (rules) respect the predefined range. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising: receiving machine-learning-based controller agent time series data describing operating states of the manufacturing equipment; and responsive to the operating states indicating the value of the at least one operating parameter is outside the predefined range, generating by the machine-learning-based controller agent a command for the manufacturing equipment to execute at least one of the control actions such that the manufacturing equipment performs the at least one of the control actions as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 19,
Gadre et al. in view of Risbeck et al. teaches the method of claim 17.
In addition, Risbeck et al. further teaches wherein the one or more rules are obtained from a rules controller (Fig. 6; [0124]-[0125]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202 … The model predictive control process performed by model predictive controller 622 may include optimizing an objective function J subject to a set of constraints. The objective function J may quantify an amount of energy consumption, a cost of energy consumption, one or more demand charges resulting from the energy consumption, penalty costs (e.g., for violating temperature bounds, for changing equipment loads rapidly, etc.), or any other cost associated with operating HVAC equipment 206 (e.g., equipment degradation, equipment purchase costs, etc.). Model predictive controller 622 may optimize the objective function J by adjusting a set of decision variables. In the context of controlling HVAC equipment 206, the decision variables may include the amount of heating or cooling
Q
˙
HVAC generated by HVAC equipment 206 and applied to building zone 202 at each of a plurality of time steps during a given time period (i.e., the optimization period). The constraints may be based on physics model 412 and may ensure that the predicted states of building zone 202 (i.e., Tz and Tm) respect the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 (rules controller) generates and imposes a set of constraints (rules)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the one or more rules are obtained from a rules controller as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Regarding Claim 20,
Gadre et al. in view of Risbeck et al. teaches the method of claim 16.
In addition, Risbeck et al. further teaches further comprising operating the manufacturing equipment with a rules controller to generate the training data set (Fig. 6; [0124]: "Model predictive controller 622 may be configured to formulate and execute a model predictive control process using physics model 412 and the values of the heat load disturbance
Q
˙
other and/or the values of the functions F(⋅), G(⋅), and H(⋅) provided by disturbance predictor 610. Model predictive controller 622 may use physics model 612 to predict the dynamic temperature response of building zone 202 (i.e., predict how the states Tz and Tm of building zone 202 will change) as a function of the amount of heating or cooling HVAC applied to building zone 202, as well as the inputs provided by disturbance predictor 610, at each of a plurality of time steps during a time period. Specifically, model predictive controller 622 may use physics model 612 to generate and impose constraints on the model predictive control process to ensure that the evolution of the states Tz and Tm follows the heat transfer dynamics of building zone 202" teaches that model predictive controller 622 (rules controller) uses (generates commands for) the physics model to generate and impose constraints (rules) on the model predictive control process such that the physics model predicts the output dynamic temperature response (e.g. temperature Tz is predicted) in response to the provided inputs for the HVAC equipment (manufacturing equipment). Fig. 6; [0098]: "a set of training data that includes values of the ambient temperature Ta, the weather forecast (e.g., cloudiness C), the time and date t, the electric load of building zone 202, measurements of the zone temperature Tz, and values of the heating or cooling duty
Q
˙
HVAC. All of these variables can be readily measured, observed, or predicted and may be available for use as training data" teaches that the predicted measurement of the temperature Tz may be used in the training data set. [0113]: "control system 600 and controller 204 can be used to control any type of environmental condition (e.g., humidity, temperature, air quality, carbon dioxide levels, pollutant levels, air pressure, air flow, lighting, etc.) within any type of space (e.g., a building zone; a vehicle such as an airplane, automobile, train, etc.; an indoor or outdoor space; a factory, etc.). Additionally, control system 600 and controller 204 can be used to control other types of systems or processes (e.g., manufacturing processes, industrial processes, construction processes, chemical processes, etc.) that are subject to disturbances regardless of whether the system or process affects environmental conditions within a building" that the controller can be used to control manufacturing processes (e.g. generate commands for operating the manufacturing equipment)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate further comprising operating the manufacturing equipment with a rules controller to generate the training data set as taught by Risbeck et al. to the disclosed invention of Gadre et al.
One of ordinary skill in the art would have been motivated to make this modification because an "advantage of splitting physics model 412 and controller model 414 into separate models is that it allows physics model 412 and controller model 414 to be different types of models ... This advantage allows controller 204 to use only the linear state-space model (i.e., physics model 412) when performing the model predictive control process, which reduces computation time and uses fewer processing resources" (Risbeck et al. [0063]).
Claims 12-14 are rejected under 35 U.S.C. 103 as being unpatentable over Gadre et al. (US 2023/0195061 A1) in view of Risbeck et al. (US 2021/0041127 A1) and further in view of Torrado et al. (US 2022/0414429 A1).
Regarding Claim 12,
Gadre et al. in view of Risbeck et al. teaches the method of claim 1.
Gadre et al. in view of Risbeck et al. does not appear to explicitly teach wherein the physics model is a sequence to sequence machine learning model.
However, Torrado et al. teaches wherein the physics model is a sequence to sequence machine learning model (Fig. 1; Fig. 5; [0089]: "The encoder 16 and the decoder 18 may be trained at the same time with physical attention and physical loss function, as shown in FIG. 5. FIG. 5 illustrates that, according to embodiments, the PIANN system 10 may be used to provide a surrogate model 200 for a physics based simulation where the surrogate model 200 is constructed using a PIANN architecture as described herein and may be considered as including or implementing the PIANN 12" teaches that the physics-informed attention-based neural network (PIANN) is a physics based model that includes an encoder and decoder. [0038]: "the PIANN includes an encoder-decoder recurrent neural network (RNN) configuration having a plurality of RNN units" teaches that the PIANN (physics model) is an encoder-decoder recurrent neural network (sequence to sequence machine learning model)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
Torrado et al. is analogous to the claimed invention because it is directed towards the implementation of a machine learning physics based model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the physics model is a sequence to sequence machine learning model as taught by Torrado et al. to the disclosed invention of Gadre et al. in view of Risbeck et al.
One of ordinary skill in the art would have been motivated to make this modification to "enable determination of the most relevant information to adapt the behavior of the neural network (RNN units) so that it generates an approximate or prediction without drawbacks noted above, including without needing residual regularization or a priori knowledge" ( et al. [0064]).
Regarding Claim 13,
Gadre et al. in view of Risbeck et al. and further in view of Torrado et al. teaches the method of claim 12.
In addition, Torrado et al. further teaches wherein the sequence to sequence machine learning model is an encoder-decoder model ([0038]: "the PIANN includes an encoder-decoder recurrent neural network (RNN) configuration having a plurality of RNN units" teaches that the PIANN (physics model) is an encoder-decoder recurrent neural network (encoder-decoder model)).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
Torrado et al. is analogous to the claimed invention because it is directed towards the implementation of a machine learning physics based model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the sequence to sequence machine learning model is an encoder-decoder model as taught by Torrado et al. to the disclosed invention of Gadre et al. in view of Risbeck et al.
One of ordinary skill in the art would have been motivated to make this modification to "enable determination of the most relevant information to adapt the behavior of the neural network (RNN units) so that it generates an approximate or prediction without drawbacks noted above, including without needing residual regularization or a priori knowledge" ( et al. [0064]).
Regarding Claim 14,
Gadre et al. in view of Risbeck et al. and further in view of Torrado et al. teaches the method of claim 13.
In addition, Torrado et al. further teaches wherein the encoder-decoder model includes long short-term memory models ([0038]: "the PIANN includes an encoder-decoder recurrent neural network (RNN) configuration having a plurality of RNN units" teaches that the PIANN (physics model) is an encoder-decoder recurrent neural network (encoder-decoder model) with a plurality of RNN units. [0021]: "the plurality of RNN units include at least one gated recurrent unit (GRU) and/or at least one long short-term memory (LSTM)" teaches that the encoder-decoder model includes long short-term memory models).
Gadre et al. and Risbeck et al. are analogous to the claimed invention because they are directed towards the use of a machine learning physics model in a manufacturing process for manufacturing equipment.
Torrado et al. is analogous to the claimed invention because it is directed towards the implementation of a machine learning physics based model.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate wherein the encoder-decoder model includes long short-term memory models as taught by Torrado et al. to the disclosed invention of Gadre et al. in view of Risbeck et al.
One of ordinary skill in the art would have been motivated to make this modification to "enable determination of the most relevant information to adapt the behavior of the neural network (RNN units) so that it generates an approximate or prediction without drawbacks noted above, including without needing residual regularization or a priori knowledge" ( et al. [0064]).
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to BRIAN J HALES whose telephone number is (571)272-0878. The examiner can normally be reached M-F 9:00am - 5:00pm.
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, Kamran Afshar can be reached at (571) 272-7796. 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.
/BRIAN J HALES/Examiner, Art Unit 2125
/KAMRAN AFSHAR/Supervisory Patent Examiner, Art Unit 2125