System and method for operation optimization of an equipment

Patent metadata

Field	Value
Publication number	US-12039434-B2
Application number	US-201917309432-A
Country	US
Kind code	B2
Filing date	Nov 28, 2019
Priority date	Nov 28, 2018
Publication date	Jul 16, 2024
Grant date	Jul 16, 2024

Abstract

This disclosure relates to optimizing an operation of an equipment by a neural network based optimizer is provided. The method include receiving, information associated with at least one equipment instance (j) as an input at a predefined sequence of timestamps; training, a plurality of simulation models for each equipment instance to obtain a function (f j ); processing, the external input parameters (e t ) to obtain a fixed-dimensional vector and passed as an input to obtain an vector (i t ); generating, a modified (i t ) from the output vector (i t ) based on a domain constraint value; computing, a reward (r t ) based on (i) the function (f j ), (ii) the modified (i t ), (iii) the external input parameters (e t ), and (iv) a reward function (R j ); and iteratively performing the steps of processing, generating, and computing reward (r t ) for a series of subsequent equipment instances after expiry of the predefined sequence of timestamps associated with a first equipment instance.

First claim

The invention claimed is:

1. A processor implemented method of optimizing an operation of an equipment by a neural network based optimizer, comprising:
receiving, information associated with at least one equipment instance (j) as an input at a predefined sequence of timestamps, wherein the information associated with the at least one equipment instance (j) corresponds to at least one of (i) a plurality of control parameters (i t ), (ii) external input parameters (e t ), (iii) output parameters (o t ), and (iv) historical operational data;
training, by a simulation model trainer, a plurality of simulation models for each equipment instance using the information associated with the at least one equipment instance (j) to obtain a function (f j ), wherein the function (f j ) corresponds to a set of simulation models of the equipment instance (j);
generating, by a sequence generator, at least one sequence of the external input parameters (e t );
processing, by a multi-head input handler, the external input parameters (e t ) to obtain a fixed-dimensional vector that is passed as an input to a neural network to obtain an vector (i t );
generating, by a domain constraints handler, a modified (i t ) from the output vector (i t ) based on a domain constraint value;
computing, by a multi-head output handler, a reward (r t ) based on (i) the function (f j ), (ii) the modified (i t ), (iii) the external input parameters (e t ), and (iv) a reward function (R j ); and
iteratively performing the steps of processing, generating, and computing reward (r t ) for a series of subsequent equipment instances after expiry of the predefined sequence of timestamps associated with a first equipment instance.

2. The method as claimed in claim 1 , wherein the multi-head input handler comprises a processing head for each equipment instance, wherein number of external input parameters differs across different instances of equipments.

3. The method as claimed in claim 1 , wherein the function f j is generated by at least one of (i) a Gaussian processes, or (ii) Gaussian mixture models constrained on dimensionality of the output vector it and the vector of external input parameters e t , and combination thereof.

4. The method as claimed in claim 1 , further comprising, inputting, the output vector (i t ) and reward (r t ) into the Neural Network with the vector of external input parameters at time (‘t+1’), e t+1 for iteratively performing the steps of processing, generating, and computing reward to obtain the set of control input parameters at time (t+1), i t+1 , wherein dimensionality of i t and e t is determined by a target domain.

5. The method as claimed in claim 1 , wherein the step of training the neural network with a loss function is performed based on a maximum attainable value of the reward function (R j ).

6. The method as claimed in claim 1 , wherein the neural network is trained by including an additional loss function using a plurality of penalty values for violating domain constraints.

7. A system for operation optimization of an equipment by a neural network based optimizer, wherein the system comprising:
a memory storing instructions;
one or more communication interfaces; and
one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
receive, information associated with at least one equipment instance (j) as an input at a predefined sequence of timestamps, wherein the information associated with the at least one equipment instance (j) corresponds to at least one of (i) a plurality of control parameters (i t ), (ii) external input parameters (e t ), (iii) output parameters (o t ), and (iv) historical operational data;
train, a plurality of simulation models for each equipment instance using the information associated with the at least one equipment instance (j) to obtain a function (f j ), wherein the function (f j ) corresponds to a set of simulation models of the equipment instance (j);
generate, at least one sequence of the external input parameters (e t );
process, the external input parameters (e t ) to obtain a fixed-dimensional vector that is passed as an input to a neural network to obtain an output vector (i t );
generate, a modified (i t ) from the output vector (i t ) based on a domain constraint value;
compute, a reward (r t ) based on (i) the function (f j ), (ii) the modified (i t ), (iii) the external input parameters (e t ), and (iv) a reward function (R j ); and
iteratively perform the steps of process, generate, and compute reward (r t ) for a series of subsequent equipment instances after expiry of the predefined sequence of timestamps associated with a first equipment instance.

8. The system as claimed in claim 7 , wherein a multi-head input handler comprises a processing head for each equipment instance, wherein number of external input parameters differs across different instances of equipments.

9. The system as claimed in claim 7 , wherein the function f j is generated by at least one of (i) a Gaussian processes, or (ii) Gaussian mixture models constrained on dimensionality of the output vector it and the vector of external input parameters e t , and combination thereof.

10. The system as claimed in claim 7 , wherein the one or more hardware processors are further configured by the instructions to input, the output vector (i t ) and reward (r t ) into the Neural Network with the vector of external input parameters at time (‘t+1’), e t+1 to iteratively perform the steps of process, generate, and compute reward (r t ) to obtain the set of control input parameters at time (t+1), i t+1 , wherein dimensionality of i t and e t is determined by a target domain.

11. The system as claimed in claim 7 , wherein the step of training the neural network with a loss function is performed based on a maximum attainable value of the reward function (R j ).

12. The system as claimed in claim 7 , wherein the neural network ( 208 A) is trained by including an additional loss function using a plurality of penalty values for violating domain constraints.

13. One or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors causes:
receiving, information associated with at least one equipment instance (j) as an input at a predefined sequence of timestamps, wherein the information associated with the at least one equipment instance (j) corresponds to at least one of (i) a plurality of control parameters (i t ), (ii) external input parameters (e t ), (iii) output parameters (o t ), and (iv) historical operational data;
training, a plurality of simulation models for each equipment instance using the information associated with the at least one equipment instance (j) to obtain a function (f j ), wherein the function (f j ) corresponds to a set of simulation models of the equipment instance (j);
generating, at least one sequence of the external input parameters (e t );
processing, the external input parameters (e t ) to obtain a fixed-dimensional vector that is passed as an input to a neural network to obtain an vector (i t );
generating, a modified (i t ) from the output vector (i t ) based on a domain constraint value;
computing, a reward (r t ) based on (i) the function (f j ), (ii) the modified (i t ), (iii) the external input parameters (e t ), and (iv) a reward function (R j ); and
iteratively performing the steps of processing, generating, and computing reward (r t ) for a series of subsequent equipment instances after expiry of th

Assignees

• Tata Consultancy Services Ltd

Inventors

• Tankasala Veparala Vishnu
• Manuelraj Solomon Pushparaj
• Bansal Ankit
• Malhotra Pankaj
• VIG Lovekesh
• Shroff Gautam
• Runkana Venkataramana
• Subramanian Sivakumar
• Pareek Aditya
• Masampally Vishnu Swaroopji
• Raj Nishit

Classifications

• G06N3/10

  Interfaces, programming languages or software development kits, e.g. for simulating neural networks · CPC title

• G06N3/0985

  Hyperparameter optimisation; Meta-learning; Learning-to-learn · CPC title

• G06N3/092

  Reinforcement learning · CPC title

• G06N3/0442

  characterised by memory or gating, e.g. long short-term memory [LSTM] or gated recurrent units [GRU] · CPC title

• G06N3/063 Primary

  using electronic means · CPC title