Neural Network Based Induction Motor Command Filtering Backstepping Control

Induction motors(IMs)have played more and more important role in industrial applications because of their simple structure,low manufacturing cost,high reliability and effectiveness.However,since the IMs’ drive system is a high-order,strongly coupling and parameter time-varying nonlinear system,which will easily reduce the IMs running efficiency under the influence of load disturbance and parameter time-varying.Therefore,the research of advanced and effective control strategy to improve the dynamic and static performance of IMs drive system is a valuable research direction in theoretical and practical applications.In order to overcome the drawbacks of the classical control method of IMs,in this paper,the neural network adaptive speed regulation control and position tracking control strategy of IMs are studied based on command filtered adaptive backstepping theorem.During the design process,neural network systems are used to approximate the nonlinear functions in the system,and by using command filtering technique and adaptive backstepping method to achieve the quick and effective control of IMs drive system.The main efforts of this paper are as follows:Firstly,the adaptive command filtered backstepping control method is developed for the strict feedback nonlinear systems.The systems’ nonlinear functions are approximated by radial basis function(RBF)neural network and the command filters are introduced to avoid the problem of “explosion of complexity”.Moreover,backstepping algorithm is applied to construct the adaptive neural network controllers and Lyapunov theory is used to prove the stability of the IMs’ drive systems.Secondly,based on command filtered backstepping technique and neural network approximation theorem,this paper designed the real controllers to solve the speed regulation control and position tracking control problems.Neural networks are used to approximate the unknown nonlinear functions in the IM drive systems and the command filters are introduced to avoid the problem of “explosion of complexity” caused by the virtual controllers.Then backstepping algorithm is applied to design the real controllers.Finally,Lyapunov theorem is proposed to analyze the stability of the system.The designed controllers only contains one adaptive parameter and can be easily implemented in real applications.Futhermore,the influence of system parameter uncertainties and load disturbances can be well handed,all the closed-loop signals are bounded.The MATLAB simulation results are provided to prove the effectiveness of the proposed control method.Thirdly,based on error compensation mechanism,the command filtered adaptive neural network speed regulation controllers are designed for the IM drive system.The nonlinear functions of the system are approximated by RBF neural network systems,the command filters are used to solve the problem of the greatly increasing computational burden due to the repeated derivation of the virtual control signals.Error compensation signals are introduced to eliminate the influence to the controlled system caused by command filters,and the backstepping technique is used to construct the real controllers of the system.The designed controllers ensure that the induction motor’s rotor speed can quickly and accurately track the given signals and all the signals are bounded.Finally,the simulation results are displayed to prove that even under the influence of parameter uncertainties and load torque disturbance,the proposed control method can grantee the great robustness and tracking performance of the system.