Fault estimation and self-healing control of discrete-time T-S fuzzy model with sensor and actuator faults based on dual observers

• Published in: Journal of process control 2023-10, Vol.130, p.103070, Article 103070
• Main Authors: Pan, Hongguang, Yu, Xinyu, She, Yangyang, Teng, Bingyang, Li, Li, Hu, Jianchen
• Format: Article
• Language: English
• Subjects: Discrete-time T-S fuzzy model; Dual observer; Fault estimation; Self-healing control; Sensor and actuator faults; Wastewater treatment process
• ISSN: 0959-1524; 1873-2771
• DOI: 10.1016/j.jprocont.2023.103070

This paper proposes a sensor and actuator fault estimation and self-healing control scheme to address issues related to deteriorating effluent quality, process failures, and safety hazards. Firstly, a discrete T-S fuzzy model of the wastewater treatment system is developed by employing the weighted recursive least squares method, which involves identifying the system based on the Benchmark Simulation Model No. 1. Secondly, in order to overcome the conservativeness problem when estimating sensor and actuator faults with a single observer, a fault estimation method employing dual observers is proposed utilizing the discrete-time T-S fuzzy model established. Thirdly, a fuzzy weighted self-healing controller is designed based on integral sliding mode and state feedback. This controller aims to overcome the aforementioned issues. Finally, Some simulations are provided to ensure the effectiveness of the obtained results. The simulation results show that compared with other methods, the fault estimation method using dual observers achieves shorter error estimation correction time and faster response to sudden fault estimation. Compared with output feedback controller, dynamic output feedback based self-healing controller exhibits superior control performance and better dynamic tracking performance. •Dual observers are proposed, which serve to simplify system modeling and mitigate conservatism in LMI-based design.•An Adaptive Kalman observer is developed, which enhances adaptability, thereby enabling quicker response times in the fault estimation process.•Self-healing controller is designed, which leverage descriptor and adaptive Kalman observers. It is outperformed by dynamic output feedback in terms of response speed and accuracy, in the context of tracking dissolved oxygen set points.