Data-driven set-point learning control with ESO and RBFNN for nonlinear batch processes subject to nonrepetitive uncertainties

Naseem Ahmad; Shoulin Hao; Tao Liu; Yihui Gong; Qing-Guo Wang

doi:10.1016/j.isatra.2023.12.044

Data-driven set-point learning control with ESO and RBFNN for nonlinear batch processes subject to nonrepetitive uncertainties

ISA Trans. 2024 Mar:146:308-318. doi: 10.1016/j.isatra.2023.12.044. Epub 2024 Jan 4.

Authors

Naseem Ahmad¹, Shoulin Hao², Tao Liu³, Yihui Gong¹, Qing-Guo Wang⁴

Affiliations

¹ Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China; School of Control Science and Engineering, Dalian University of Technology, Dalian 116024, China.
² Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China; School of Control Science and Engineering, Dalian University of Technology, Dalian 116024, China. Electronic address: slhao@dlut.edu.cn.
³ Key Laboratory of Intelligent Control and Optimization for Industrial Equipment of Ministry of Education, Dalian University of Technology, Dalian 116024, China; School of Control Science and Engineering, Dalian University of Technology, Dalian 116024, China. Electronic address: liurouter@ieee.org.
⁴ Institute of Artificial Intelligence and Future Networks, Beijing Normal University at Zhuhai, Zhuhai, China; BNU-HKBU United International College, Tangjiawan, Rd. JinTong 2000#, Zhuhai, China.

PMID: 38199841
DOI: 10.1016/j.isatra.2023.12.044

Abstract

This paper proposes an extended state observer (ESO) based data-driven set-point learning control (DDSPLC) scheme for a class of nonlinear batch processes with a priori P-type feedback control structure subject to nonrepetitive uncertainties, by only using the process input and output data available in practice. Firstly, the unknown process dynamics is equivalently transformed into an iterative dynamic linearization data model (IDLDM) with a residual term. A radial basis function neural network is adopted to estimate the pseudo partial derivative information related to IDLDM, and meanwhile, a data-driven iterative ESO is constructed to estimate the unknown residual term along the batch direction. Then, an adaptive set-point learning control law is designed to merely regulate the set-point command of the closed-loop control structure for realizing batch optimization. Robust convergence of the output tracking error along the batch direction is rigorously analyzed by using the contraction mapping approach and mathematical induction. Finally, two illustrative examples from the literature are used to validate the effectiveness and advantage of the proposed design.

Keywords: Dynamic linearization data model; Extended state observer; Nonlinear batch processes; Nonrepetitive uncertainties; Radial basis function neural network; Set-point learning control.