System and method for optimization of industrial processes

Patent metadata

Field	Value
Publication number	US-2021303742-A1
Application number	US-202117178042-A
Country	US
Kind code	A1
Filing date	Feb 17, 2021
Priority date	Mar 26, 2020
Publication date	Sep 30, 2021
Grant date	—

Abstract

This disclosure relates generally to system and method for optimization of industrial processes, for example a tundish process. Typically geometries for industrial processes are simulated in a numerical analysis model such as a CFD. In order to simulate a physical phenomenon (such as tundish process) numerically, the domain thereof is discretized in order to convert the differential equations to be solved in the domain into linear equations. The accuracy of a CFD solution is dependent on a mesh of the domain, which in turn depends on a geometry thereof. For setting up an optimization task, the disclosed method provides first a CFD friendly base geometry, so that a faulty geometry can be detected before forming the complete geometry.

First claim

What is claimed is:

1 . A processor-implemented method comprising:
receiving a plurality of model parameters associated with an industrial process, via one or more hardware processors, the plurality of model parameters comprising a plurality of geometrical parameters and a plurality of operational parameters of an equipment and an optimization function associated with an optimization of the industrial process; generating, via the one or more hardware processors, a first objective function data for an objective function associated with the optimization task by using one or more model parameters from amongst the plurality of model parameters, wherein generating the first objective function data comprises:
iteratively and selectively varying the plurality of model parameters to obtain a data set of the model parameters,
checking the plurality of geometrical parameters associated with the data set based on a set of rules to identify a set of geometrical parameters capable of facilitating creation of valid geometries for a numerical analysis model, and
performing simulations using at least the set of geometrical parameters as model parameters on the numerical analysis model to generate the first objective function data;
 identifying, via the one or more hardware processors, a set of significant variables from amongst the plurality of model parameters by analyzing relative variation in the first objective function data with respect to model parameters;  creating, via the one or more hardware processors, a Design of Experiments (DOE) table comprising model parameters and variations of model parameters based on the set of significant variables by using a DOE table creation model;  generating, via the one or more hardware processors, a second objective function data using the DOE table; creating, via the one or more hardware processors, a surrogate model capable of predicting objective function data corresponding to the set of significant parameters using the second objective function data and a surrogate modelling technique; and optimizing, via the one or more hardware processors, the industrial process using the surrogate model and an optimization model.

2 . The method of claim 1 , wherein the DOE table is created by using one of a Latin Hypercube model, Full Factorial model, and Taguchi model.

3 . The method of claim 1 , wherein the surrogate modelling technique comprises one of a linear model, polynomial model, K-Nearest Neighbor (kNN) model, and a Random Forest model.

4 . The method of claim 1 , wherein the optimization model comprises one of a genetic algorithm (GA) and gradient based method.

5 . The method of claim 1 , wherein the numerical analysis model comprises a CFD model.

6 . The method of claim 1 , wherein a rule from amongst the set of rules comprises a rule associated with non-intersecting lines and surfaces associated with the geometry of the equipment.

7 . The method of claim 1 , wherein a rule from amongst the set of rules comprises a rule associated with formation of closed geometry of the equipment.

8 . A system ( 100 ) comprising:
one or more memories ( 106 ); and one or more hardware processors ( 102 ), the one or more memories ( 106 ) coupled to the one or more hardware processors ( 102 ), wherein the one or more hardware processors ( 102 ) are configured to execute programmed instructions stored in the one or more memories ( 106 ), to: receive a plurality of model parameters associated with an industrial process, the plurality of model parameters comprising a plurality of geometrical parameters and a plurality of operational parameters of an equipment and an optimization function associated with an optimization of the industrial process; generate a first objective function data for an objective function associated with the optimization task by using one or more model parameters from amongst the plurality of model parameters, wherein generating the first objective function data comprises:
iteratively and selectively vary the plurality of model parameters to obtain a data set of the model parameters,
check the plurality of geometrical parameters associated with the data set based on a set of rules to identify a set of geometrical parameters capable of facilitating creation of valid geometries for the numerical analysis model, and
perform simulations using at least the set of geometrical parameters as model parameters on the numerical analysis model to generate the first objective function data;
identify a set of significant variables from amongst the plurality of model parameters by analyzing relative variation in the first objective function data with respect to model parameters; create a Design of Experiments (DOE) table comprising model parameters and variations of model parameters based on the set of significant variables by using a DOE table creation model; generate, via the one or more hardware processors, a second objective function data using the DOE table; create a surrogate model capable of predicting objective function data corresponding to the set of significant parameters using the second objective function data and a surrogate modelling technique; and optimize the industrial process using the surrogate model and an optimization model.

9 . The system of claim 8 , wherein the DOE table is created by using one of a Latin Hypercube model, Full Factorial model, and Taguchi model.

10 . The system of claim 8 , wherein the surrogate modelling technique comprises one of a linear model, polynomial model, K-Nearest Neighbor (kNN) model, and a Random Forest model.

11 . The system of claim 8 , wherein the optimization model comprises one of a genetic algorithm (GA) and gradient based method.

12 . The system of claim 8 , wherein the numerical analysis model comprises a CFD model.

13 . The system of claim 8 , wherein a rule from amongst the set of rules comprises a rule associated with non-intersecting lines and surfaces associated with the geometry of the equipment.

14 . The system of claim 8 , wherein a rule from amongst the set of rules comprises a rule associated with formation of closed geometry of the equipment.

15 . One or more non-transitory machine readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:
receiving a plurality of model parameters associated with an industrial process, via one or more hardware processors, the plurality of model parameters comprising a plurality of geometrical parameters and a plurality of operational parameters of an equipment and an optimization function associated with an optimization of the industrial process;
generating, via the one or more hardware processors, a first objective function data for an objective function associated with the optimization task by using one or more model parameters from amongst the plurality of model parameters, wherein generating the first objective function data comprises:
iteratively and selectively varying the plurality of model parameters to obtain a data set of the model parameters,
checking the plurality of geometrical parameters associated with the data set based on a set of rules to identify a set of geometrical parameters capable of facilitating creation of valid geometries for a numerical analysis model, and
performing simulations using at least the set of geometrical parameters as model parameters on the numerical analysis model to generate the first objective function data;

identifying, via the one or more hardware processors, a set of significant variables from amongst the plurality of

Assignees

• Tata Consultancy Services Ltd

Inventors

• Singh Prateek
• Ahmad Dilshad
• MANGAL Saurabh

Classifications

• G06F30/27

  using machine learning, e.g. artificial intelligence, neural networks, support vector machines [SVM] or training a model · CPC title

• G05B17/02 Primary

  electric · CPC title

• G06F30/10 Primary

  Geometric CAD · CPC title

• G05B13/042

  in which a parameter or coefficient is automatically adjusted to optimise the performance · CPC title

• G06N3/126

  Evolutionary algorithms, e.g. genetic algorithms or genetic programming · CPC title