Production Process Control with Dynamic Regression Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing regression-optimization methods in manufacturing systems fail to accurately model and optimize production processes with dynamic inputs, particularly when underlying physical and chemical relationships are complex and non-linear, and do not account for multiplication of control variables with observed variables.
Innovation Solution
A computer-implemented method using a physics-based expression to determine coefficient and bias terms for a dynamic linear model, which is integrated into an optimization model to generate a regression-optimization model that dynamically adjusts control variables based on changing input variables, employing techniques like mixed integer linear programming and gradient-based optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional regression-optimization methods are used to model production processes, then the model structure is simple and easy to implement, but the model cannot accurately capture complex non-linear physical and chemical relationships
Solution Approach 1:
The patent applies dynamics by transforming static regression coefficients into dynamic expressions that adapt to changing operating conditions. The coefficient generation module creates coefficients as functions of observed variables, allowing the model to dynamically adjust to non-linear relationships in production processes while maintaining the linear model structure.
Solution Approach 2:
The patent changes parameters by allowing regression coefficients to vary based on observed variables rather than remaining constant. This is achieved through the coefficient generation module that computes coefficients as dynamic expressions involving observed variables, enabling the model to capture non-linear relationships without changing the fundamental linear model form.
2Adaptability or versatility
If dynamic linear models with coefficient expressions are used, then the model can adapt to changing input variables, but the computational complexity increases
Solution Approach 1:
The patent applies preliminary action by pre-generating coefficient expressions in terms of observed variables before optimization. The coefficient generation module creates these dynamic coefficient expressions in advance, which are then directly substituted into the optimization model, avoiding the need for complex real-time computations during the optimization process.
3Manufacturing precision
If real-time optimization is performed with complex non-linear relationships, then the control accuracy is improved, but the processing time increases
Solution Approach 1:
The patent extracts the non-linear relationships from the optimization problem by representing them as pre-generated coefficient expressions in the linear optimization model. This separation allows the optimization solver to work with a linear model structure while the coefficient expressions capture the non-linear physics and chemistry, significantly reducing processing time compared to directly optimizing non-linear models.
Data Source
AI summary
Dynamic control of a production process of a manufacturing system is facilitated, where the control process includes receiving runtime input data for multiple input variables of the production process. The production process is represented, at least in part, by a physics-based expression, with at least one term of the physics-based expression being a function of two or more input variables of the production process. The control process includes determining coefficient and bias terms for a dynamic linear model connecting the multiple input variables and an output of the production process, where the terms are based, at least in part, on the input variables. The dynamic linear model and determined coefficient and bias terms are provided in an optimization model to generate a regression-optimization model which determines an optimized value of a control variable for the production process, which is used in facilitating control of the production process.