A reconfigurable PID fault tolerant tracking controller design for LPV systems. (March 2020)

• Record Type: Journal Article
• Authors: Rabaoui, B.; Hamdi, H.; Braiek, N. BenHadj; Rodrigues, M.
• Is Part Of: ISA transactions. Volume 98(2020)
• Page Start: 173
• Page End: 185
• Publication Date: 2020-03
• Subjects: LPV system -- PID fault tolerant tracking controller -- Adaptive polytopic observer -- LMIs; Engineering instruments -- Periodicals; Engineering instruments; Periodicals; Electronic journals; 629.805
• DOI: 10.1016/j.isatra.2019.08.049
• Languages: English
• ISSNs: 0019-0578

Abstract

Abstract: This paper considers the design of a reconfigurable PID Fault Tolerant Tracking Controller (PID-FTTC) for Linear Parameter Varying (LPV) systems affected by actuator faults with the presence of disturbance. The LPV systems are represented through a polytopic LPV description with measurable gain scheduling functions. A new PID-FTTC scheme with a model reference, an adaptive PID controller and an Adaptive Polytopic Observer (APO), is developed. The main idea is to improve and to compare performances with this developed PID-FTTC versus previous similar FTC techniques especially about the settling time, the overshoot and integral error indices. By the way, this paper can reduce the conservatism of previous methods with more parameters design so as to avoid their disadvantages and to give better control loop performances especially in terms of accuracy and speed of trajectory tracking even when a fault occurs. So, in order to establish the stability of the reconfigured PID-FTTC, a new theoretical study is developed through the use of Linear Matrix Inequality (LMI). This new method is illustrated through a two-tank process where the results compared to previous ones, underline the improvements. Highlights: A new PID-FTTC scheme with a model reference, an adaptive PID controller and an Adaptive Polytopic Observer (APO), is developed. Based on these on-line informations, the PID-FTTC is reconfigured to compensate fault effects, to ensure the stability, the accuracy and the reference outputs tracking with quick response. This new method is illustrated through a two-tank process where the results are compared to previous ones and underline the improvements.