Fault-Tolerant Control For Nonlinear Continuous Processes And Batch Processes

Since fault-tolerant control can improve security and reliability of systems and guarantee the closed-loop performance in faulty cases, it has been widely studied in academic and industrial fields. Up to the present, research on fault-tolerant control has lasted more than thirty years. Considering all fault-tolerant control systems as a set, we can analyze this set from three different points of views: according to the system characteristics, it can be divided into linear systems and nonlinear systems; according to the class of used methods, it contains passive and active methods; according to the manner of manufacture, it has continuous and batch processes.There are plenty of reported works on passive and active fault-tolerant control methods for linear continuous processes; passive methods for nonlinear continuous processes have some big shortcomings; while fault-tolerant control for batch processes is an open problem; therefore, this dissertation focuses on active fault-tolerant control of nonlinear continuous processes and fault-tolerant control of batch processes.Active fault-tolerant control for nonlinear continuous processes is studied in Chapters 2-5. In Chapter 2, the identifiability of sensor faults is studied for a class of nonlinear systems, then some fault detection and diagnosis methods are presented, in addition, the above results are used to design an active fault-tolerant control scheme. Furthermore, uncertainties especially unmatched uncertainties are studied in Chapters 3 and 4. To compensate unmatched uncertainties, a variable structure control law is proposed for a class of second-order nonlinear systems in Chapter 3. By using the scheme proposed in Chapter 3, Chapter 4 provides an active fault-tolerant control scheme for a class of nonlinear non-minimum phase systems with component faults and unmatched uncertainties. A novel fault form, actuator stiction fault, is described in Chapter 5, and an improved Generalized Predictive Control law is proposed to deal with this kind of faults.Chapters 6-8 focus on fault-tolerant control for batch processes. To the best knowledge of the authors, these works are the inaugurator in this field. Chapter 6 presents an iterative learning reliable control law for a class of batch processes with actuator faults, which is based on static state feedback. The method used in Chapter 7 is similar to that used in Chapter 6, but the considered faults are sensor faults and the control law is based on dynamic output feedback. In Chapter 8, an active fault-tolerant control method is proposed for a class of nonlinear batch processes with sensor faults. The proposed method has been used to a Chen Hsong injection molding machine and the experimental results are excellent.