Finite-time Fuzzy Adaptive Control Of Complex Nonlinear Systems

With the continuous development of science and technology and the advent of the era of artificial intelligence,the research on intelligent control is becoming more and more important.Intelligent control is an advanced stage of the development of control science and technology.It can solve the problem that traditional methods are difficult to solve by learning the laws of human intelligent activities and information transmission processes without requiring an accurate model.The control goal often needs to be completed within a finite time,and the actual control system needs to meet the condition of finite-time stability.Stability is an important part of the study of dynamic systems.The traditional Lyapunov function characterizes the steady-state performance of the system,and the convergence form of the closed-loop system limits the better control performance.These problems have prompted the research on finite-time stability control theory.The traditional stability is asymptotic stability.Asymptotic stability studies the stability of the system when the time tends to infinity.The finite time stability emphasizes that the state trajectory stays within the prescribed bound for a certain period,and it is the timeoptimal control.This dissertation studies the problem of finite-time intelligent control design and stability analysis,and uses the intelligent control technology to deal with the unknown items of nonlinear system.Based on the Lyapunov function,the stability,convergence and robustness of the closed-loop system are given.Under the framework of the finite-time stability theory,this dissertation studies the intelligent control issues for complex nonlinear systems with periodic sampling and event-triggered mechanisms,using intelligent control technology and nonlinear control technology as tools and combining filter technique,fault-tolerant control method,prescribed performance control method,switched systems theory,etc.The main research contents are as follows:(1)The increasing complexity of the control system,the increasing control scale and the continuous improvement of the demand for control performance have made the traditional control theories and methods unable to meet the design requirements.Intelligent control does not require an accurate model of the system,and can solve complex control problems that are difficult to solve by traditional methods.The control performance of the control system will inevitably be affected by various interferences,including external environmental interference and input channel interference.This dissertation first studies the finite-time performance control problem under the periodic sampling mechanism for a class of strict feedback nonlinear systems with external disturbances.The main idea is to use a piecewise function to overcome the traditional exponential convergence rate that cannot ensure that the tracking error converges to a bounded set within a set time.The disturbance observer is designed to estimate the unknown disturbance.The control signal introduces the feedforward compensation term of disturbance estimation.The fuzzy control technology is used to deal with the unknown nonlinear term of the system.Under the framework of finite-time stability,a fuzzy adaptive finite-time control scheme based on the disturbance compensation and performance function is proposed.(2)The problem of finite-time fault-tolerant control of a class of nonlinear systems with non-affine function faults is studied.The complexity of the actual physical system increases the probability of faults of the control system.The faults may cause the quality and performance of the control to decrease,and affect the stability of the system.Therefore,it is important to design an effective control scheme to reduce the performance degradation caused by system faults,and to improve the safety and reliability of the control system.The research of fault-tolerant control provides an effective way to solve this safety problem.Different from linear faults(stuck and lose effectiveness faults,bias and gain faults,etc.),the non-affine function faults considered in this dissertation are nonlinear faults.In the control scheme,the fuzzy logic system and Butterworth low-pass filter technology are combined to solve the unknown nonlinear faults problem.The designed control strategy adopts the filter signal to process the non-affine term.The proposed fuzzy adaptive fault-tolerant finite-time control scheme under the periodic sampling mechanism can not only ensure that the closed-loop system is semi-globally practically finite-time stable,but also the tracking error converges to a small residual set.(3)The above two main contents have respectively studied the control problem with/without system faults.Considering that most control schemes in practice are applied to the control system through digital controllers,periodic sampling control(periodic sampling control)may increase the unit operation and communication load of the control process,so the issue of event-triggered control is studied next.Event-triggered control is a non-periodic control method.Its principle is to reduce the internal information transmission load of the system by designing the event-triggered rules(defining a changed signal as the trigger threshold),and effectively reduce the unit of the control process under the premise of the performance or quality.It has unique advantages in reducing control energy consumption and occupancy of transmission bandwidth.On this basis,this dissertation studies the event-triggered robust finite-time control problem a class of nonlinear systems based on exponential performance functions and relative thresholds.The relative threshold of event-trigger control reduces the communication burden and reduces unnecessary waste of communication resources,which effectively improves the unit computing capability of the control process.Based on the disturbance observer,the coupling disturbance composed of fuzzy approximation error and external interference is estimated.Because the designed controller contains the estimation signal of the coupling disturbance,the robust control performance is improved.(4)Based on the above three main research contents,considering the complex dynamic characteristics of the control system,that is,the complex relationship between the nonlinear terms and state variables,the problem of finite-time event-triggered control of non-strict feedback systems is studied.Each sub-equation of this kind of control system contains all the state variables of the system,rather than part of the state variables.This dissertation uses fuzzy logic systems to model nonlinear systems,designs the changed signal as the trigger signal,and proposes two finite-time event-triggered control schemes with/without faults.Among them,the problem of switching control with arbitrary switching signals is studied,which is different from the multi-Lyapunov function switching control method using average dwell time.In this dissertation,a common Lyapunov function is used to design a control scheme that does not rely on switching signals,which reduces the design complexity of the controller.The problem that the signal value of any switching signal is difficult to capture is solved.The proposed scheme can not only ensure the tracking control performance of the system,but also reduce the communication load of the control unit and reduce energy consumption.Finally,the conclusion of this dissertation is given,the main research content and innovation are summarized,and the shortcomings and the existing to be solved are briefly analyzed.