Event-triggered Adaptive Fault-tolerant Control For Uncertain Nonlinear Systems

As the scale and complexity of systems used in actual production applications increases,and the environment of the system becomes more unpredictable,the actual system usually consists of uncertain nonlinear systems with unknown parameters and unknown nonlinear items.In such complex and uncertain systems,faults such as actuators,sensors and components are inevitable,which will inevitably affect the control performance of the system and even lead to the destruction of system reliability and safety.It is well known that the requirements for control performance in actual systems are also higher and more precise,so the design of the system controller should also consider the restrictions of input constraints,state constraints,output constraints in the system.If these constraints are violated,the control performance may decline.In addition,the excessive occupation and waste of communication bandwidth and computing resources in the control process of the system are also directions that need attention.Therefore,this paper proposes a low-complexity adaptive fault-tolerant control method for uncertain nonlinear systems with faults and various constraints,which solves the control problems of uncertain nonlinear systems with actuator,component and sensor faults,input constraints,time-varying asymmetric state constraints and output constraints,and realizes the precise tracking of the desired trajectory,improving the control performance and reducing the occupation of communication bandwidth and resources.The controller structure is simple and the calculation is low,which is easy to apply to actual systems.The main research contents of this thesis can be summarized as follows.Firstly,a low-complexity approach to adaptive fault-tolerant tracking control is proposed for uncertain nonlinear systems that suffer from actuator and component faults,and unknown control directions.This proposed method utilizes transformation of error variables related to predetermined performance and does not require any compensators or approximate structures to handle the nonlinearity,actuator and component faults.This controller design method is distinct from the traditional backstepping method,which requires the derivation of virtual control signals in the design process,and from the methods based on neural networks,fuzzy logic systems,which need to approximate to obtain the unknown model of the system.This controller design method is advantageous as it does not require accurate model information,preventing the explosion problem in the backstepping method.It is also effective in solving external interference,unknown system uncertainty,and actuator and component faults,and can achieve system tracking control with asymptotic stability.Secondly,a novel adaptive fault-tolerant finite-time tracking control method is proposed for multi-input multi-output(MIMO)uncertain nonlinear systems with actuator and component faults.This method utilizes the measurable states to design the controller without prior knowledge of the system nonlinear functions,increasing the robustness to uncertainties and faults.A relative threshold and a fixed threshold event-triggered mechanism are established to reduce communication bandwidth and computing resources.Predetermined performance control is used to pre-specify the system tracking error,improving the tracking error performance of transient and steady-state,including overshoot,convergence speed,while ensuring the stability of the closed-loop systems.This method is able to complete the tracking control in a finite time,achieving high tracking accuracy and fast response.Then,a new low-complexity adaptive saturated fault-tolerant control approach is presented for uncertain nonlinear systems that confront with input constraints,time-varying asymmetric state constraints,and actuator faults.A bounded function with the auxiliary variable is utilized to design a new auxiliary system,thereby decreasing the negative impacts of actuator saturation on the system.This controller design method is beneficial as it can effectively handle state constraints,actuator saturation,faults,and uncertainties of unknown systems.It also simplifies the design of the backstepping method,improves the control performance of the system,and reduces communication and computing resources.As a result,the closed-loop system is ensured to be asymptotically stable.Finally,a novel low-complexity fault-tolerant tracking approach which includes asymmetric output constraints is presented for a MIMO uncertain nonlinear system taking into account the impacts of multiplicative/additive faults on sensors and actuators,and the unknown control directions.This approach implements an event-triggering mechanism in adaptive fault-tolerant control,granting it the advantage of a simple structure and a low computational complexity,resulting in an effective saving of communication and computing resources.The proposed constraint tracking control is robust and effective in solving the problem of unknown control direction and time-varying asymmetric output constraint.It eliminates the adverse effects of all state sensor faults on output tracking except for output sensor faults,and does not require any approximation technology.The contradictory Lyapunov stability analysis ensures the stability of the closed-loop system,the boundedness of all signals,the maintenance of the output constraint,and the prevention of Zeno behavior.To explore the application potential and demonstrate the effectiveness and superiority of the proposed methods,a real linear motor system platform was used to conduct a series of experiments,and the obtained results were carefully compared and analyzed.