Robust Predictive Fault-Tolerant Control for Multi-Phase Batch Processes With Interval Time-Varying Delay

Huiyuan Shi,Chengli Su,Ping Li

doi:10.1109/access.2019.2940275

Abstract

In view of multi-phase batch processes with interval time-varying delay, uncertainties, unknown disturbances, partial actuator failures and input and output constrains in real-world industrial production, a robust predictive fault-tolerant control (RPFTC) method is proposed in this paper. First, a multi-phase batch process considering the above process dynamics is described by a switching model that consists of different dimensional sub-systems. Then the switching model is transformed into the extended switching state space model by the introduction of output tracking error. On basis of this extended model, a robust predictive fault-tolerant control law is designed to improve the control performance and to obtain more degrees of freedom of the adjustment for the controller. Second, by the utilization of Lyapunov function theory, switching system theory and average dwell time approach, the sufficient conditions in terms of linear matrix inequality (LMI) constraints and minimum running time at each phase are given to make the corresponding discrete-time switching closed-loop system robustly exponential stable and the running time of each phase shortest. At the same time, the optimal cost function and H-infinity performance index are considered in the derivation of stable conditions, which can obtain the optimized control performance and suppress the unknown disturbances. Finally, the gain of the control law and the minimum running time of each phase are calculated by solving these LMIs. Taking the injection molding process as a simulation object, the control results verify the effectiveness and feasibility of the proposal.

Highlights

In order to obtain the low-volume and high-value products, researches of batch process control have attracted more and more interest from domestic and foreign scholars
For batch processes with state delay and actuator failure, Wang et al [13] proposed a twodimensional robust iterative learning fault-tolerant control (FTC) method based on Fornasini-Marchsini model
(1) The extended switching state space model including state variables and output tracking error are constructed to describe a class of multi-phase batch processes with interval time-varying delay, partial actuator failures and other problems

Summary

INTRODUCTION

In order to obtain the low-volume and high-value products, researches of batch process control have attracted more and more interest from domestic and foreign scholars. For batch processes with state delay and actuator failure, Wang et al [13] proposed a twodimensional robust iterative learning FTC method based on Fornasini-Marchsini model. On this basis, Wang et al [14] proposed iterative learning fault-tolerant guaranteed cost control method. A robust predictive fault-tolerant control method is proposed for multi-phase batch processes with interval time-varying delay, uncertainties, unknown disturbances, partial actuator failure, input and output constrains. (1) The extended switching state space model including state variables and output tracking error are constructed to describe a class of multi-phase batch processes with interval time-varying delay, partial actuator failures and other problems.

MAIN DEFINITIONS AND LEMMAS

MAIN THEOREM AND COROLLARY

Findings

CONCLUSION