Optimization Control of the Color-Coating Production Process for Model Uncertainty

Area of Science:

• Chemical Engineering
• Process Control
• Manufacturing Systems

Background:

• Color-coating production processes (CCPP) face challenges in optimizing economic efficiency and quality due to inherent model uncertainties.
• Existing control strategies often struggle to effectively manage these uncertainties, leading to suboptimal performance.

Purpose of the Study:

• To develop an advanced optimization control strategy for CCPP that explicitly addresses and mitigates model uncertainty.
• To improve both the economic efficiency and film thickness control within the CCPP.

Main Methods:

• A mechanistic model of CCPP was developed and used to generate process data.
• Partial Least Squares (PLS) was employed to create predictive models for film thickness and economic efficiency.
• Robust optimization and iterative learning control were integrated to manage model uncertainty.
• Fuzzy parameter adjustment was utilized for rapid convergence of economic efficiency and film thickness.

Main Results:

• The proposed strategy effectively manages model uncertainty in CCPP optimization.
• Simulation results demonstrated the successful refinement of model uncertainty and improved control.
• The approach ensures rapid convergence to optimized economic efficiency and film thickness under constraints.

Conclusions:

• The developed optimization control strategy provides a robust solution for CCPP, enhancing economic performance and quality.
• The integration of robust optimization, iterative learning control, and fuzzy logic offers a powerful framework for managing complex industrial processes.
• The strategy's effectiveness is validated, paving the way for practical implementation in color-coating production.