Self-healing Control And Its Application In The Networked Control Systems

With the rapid development of science and information technology,the complexity of various systems design and coupling degrees among system components have also increased.The system vulnerability becomes more prominent than before,and the fault poses a non-ignorable threat to the regular and safe operation of the system.To improve the system's ability to monitor and respond to its faults,scholars propose the thought of self-healing control.At present,self-healing control is still in the initial stage of development,and no scholars have given basic concepts such as a clear definition and research scope of self-healing control.The research on self-healing control theory in academia has just started,and the research on self-healing control theory lags behind the engineering application of self-healing control.This thesis mainly discusses the theoretical research of self-healing control,and the basic concepts such as the definition,basic structure,and research scope of self-healing control are formulated for the first time.In this thesis,the multiple uncertainties that the network control systems face are dealt with by using the self-healing control method,and the proposed self-healing control theory is verified.The main work content is as follows:First,the relevant research achievements of self-healing control for smart grids,flight control systems,and mechanical faults are reviewed.The characteristics and functions of self-healing control are clarified,and its definition is formulated.The relationship and difference between self-healing control and self-healing system as well as fault-tolerant control are analyzed.The significance of self-healing control is summarized,and its development trend is discussed.Second,a self-healing control method based on a state observer is proposed.The method is composed of a state observer method for fault diagnosis and a fault handling mechanism.Fault estimation,fault detection,and fault location are included in the designed fault diagnosis method.The amplitude information of additive faults in the system can be obtained through fault estimation.The real-time performance of fault detection and isolation is improved by refined fault detection and fault location methods.The fault information obtained by fault diagnosis is combined with the fault handling mechanism in the controller is designed so that the impact of the fault on the system is eliminated in real-time.Third,a self-healing control method based on a two-stage Kalman filter is designed.In view of the partial failure of the actuator,the two-stage Kalman filter is used to diagnose the partial failure of the actuator and the obtained fault information has high accuracy,but the real-time performance is poor.In order to improve the real-time performance of the fault diagnosis,the fault diagnosis method based on a two-stage Kalman filter and BP neural network is proposed.Moreover,online control law reconstruction/control law switching methods are combined,the impact of the fault on the system is eliminated/inhibited.At the same time,unknown failure in the system or complete failure of the actuator is considered.A self-healing control method based on fault isolation is proposed to prevent the continuous impact of faults on the system.Finally,a self-healing control scheme based on network switch is designed.Aiming at the randomness of faults or denial-of-service attack in the system,the system anomaly detection method based on the chi-square detection is designed.The selection method of the anomaly detection threshold is improved,the difficulty of the detection threshold selection is reduced.Then,the anomaly diagnosis method based on support vector machine is adopted,and the abnormal condition is classified.In the case of a denial-of-service attack in the system,the results of anomaly detection and diagnosis are combined to design a self-healing strategy based on network switch.Therefore,the network control systems can self-healing in the event of the denial-of-service attack,and the system performance can be maintained at an acceptable level.To verify the effectiveness of the self-healing control method proposed in this thesis,the MATLAB toolbox True Time is used to build a networked control system for numerical simulation.Finally,the work of this thesis is summarized,and the work that needs further research has prospected.