Fault-tolerant Control For Networked Batch Processes Based On Iterative Learning Method

As a typical high value-added,low-volume,multi-variety,multi-function and technology-intensive industrial processes,batch processes are widely used in fine chemical industry,food processing,polymeric materials,metal processing,semiconductor integrated circuit production and other fields.Due to the fact that the repetitive nature of batch processes can be effectively exploited by iterative learning method,and considering the inevitable faults problems,the fault-tolerant control based on the iterative learning method is introduced into batch processes.Network communication is a low-cost,fast and efficient way of communication,which plays an irreplaceable role in the information transmission of batch processes.However,a series of networked phenomena have a great impact on the system performance.Therefore,based on iterative learning method,the fault-tolerant control of networked batch processes is studied in this paper.The main research contents include:1.Considering the different faults(actuator faults and sensor faults)and networked phenomena(event-triggered transmission and data dropouts),after a series of deductions,the original system model is finally transformed into a two-dimensional fault-tolerant system model which can be described from two dimensions of time and batch by combining the iterative learning control method.2.Based on the dynamic output feedback,the problem of iterative learning fault-tolerant control for networked batch processes with event-triggered transmission mechanism and sector-bounded nonlinearities is studied.In this part,a new networked batch process model with sector-bounded nonlinearity is proposed by combining event-triggered transmission scheme.Furthermore,a reasonable iterative learning fault-tolerant controller is designed to ensure the asymptotic stability and robust H_?performance of the system.Finally,combined with the injection speed control process of industrial injection moulding machine,the feasibility and validity of the designed controller are verified.3.Based on the state feedback,the problem of iterative learning fault-tolerant control for batch processes with measurement packet losses is studied.Considering the characteristics of different signal transmission processes,event-triggered transmission mechanism and data packet losses are introduced in the signal transmission process of control input and measurement output,respectively.At the same time,the problem of sensor faults is also discussed in this part to fit in with the actual industrial process.By introducing a random variable satisfying Bernoulli distribution,the phenomenon of data packet losses can be described and the original system can be transformed into a two-dimensional stochastic system.Then,the mean-square asymptotic stability and the robust H_?performance of the system are analyzed on the basis of the designed controller.Finally,an example of nozzle pressure control process of industrial injection moulding is utilized to demonstrate the feasibility and validity of the designed controller.