Tolerant Control Of Nonlinear Multi-agent System Based On Wavelet Neural Network

With the development and progress of science and technology and the need of industrial production,the scale and complexity of the actual engineering system are increasing rapidly,which means that the system is more prone to failure and the failure will have a very serious impact on the system.In order to ensure the safety and reliability of the system,it is particularly important to establish an effective fault-tolerant control mechanism for the system.In recent years,multi-agent system has been widely used in various fields.Therefore,fault-tolerant control for multi-agent system is of great significance.In the early study of multi-agent system,linear system was emphasized,but the actual system is mostly non-linear system.Therefore,the current research focuses on non-linear multi-agent system with more practical significance.In this paper,fault-tolerant control of nonlinear multi-agent system is mainly studied.Because wavelet neural network has a good nonlinear approximation ability,the whole paper paper chooses the fault estimation method based on wavelet neural network to obtain the fault estimation information.First,for a class of nonlinear multi-agent system with actuator fault,the fault estimation method based on wavelet neural network is described in detail.Assuming that there is no leader in the system,the system model is established firstly;secondly,based on this model,a distributed nonlinear observer is given,that is,an observer is designed for each agent,and the state estimation information and relative output information are obtained by the observer.Then the relative output information is used as the input of the wavelet neural network,and the fault estimation information is obtained through the training of the wavelet neural network.Finally,the system model of a single link manipulator driven by two DC motors is established,and the validity of the simulation method is verified.Secondly,fault-tolerant control is applied to a class of nonlinear multi-agent system with actuator fault.Assuming that the system has a healthy leader,the system model is established first.Secondly,the fault estimation method based on wavelet neural network is used to obtain the fault estimation information,and then a distributed fault-tolerant controller is constructed to ensure the leader following consistency of the system and compensate the influence of actuator fault on the system.In addition,the Lyapunov function is used to analyze the stability of the system.Finally,the system model of the manipulator is established,which is composed of six DC motors driving the single link of the rotary joint,and the effectiveness of the method in this chapter is verified by simulation.Thirdly,fault-tolerant control is applied to a class of nonlinear multi-agent systems with actuator faults,which include external disturbances and unknown model uncertainties.Assuming that the system has a leader,the system model is established first.Secondly,the fault estimation method based on wavelet neural network is used to obtain the fault estimation information,and then a distributed collaborative fault-tolerant controller is constructed to ensure the leader following consistency of the system and compensate the influence of actuator fault,external disturbance and unknown model uncertainty on the system.In addition,the Lyapunov function is used to analyze the stability of the system.Finally,the system model of the manipulator is established,which is composed of six DC motors driving the single link of the rotary joint,and the effectiveness of the method in this chapter is verified by simulation.