Research On Adaptive Control Strategy Of Permanent Magnet Synchronous Motor Based On Neural Network And Parameter Identification

Electric vehicles reflect the current development direction of new energy vehicles.Permanent Magnet Synchronous Motor(PMSM)has been successfully applied in the field of electric vehicle motors because of its high efficiency,energy saving,reliability,large power density,and simple control,And there is huge room for development.Adaptive control is an important branch of modern control theory.It can generate feedback control rules to change the control parameters according to the detected changes in system parameters,so that the system can achieve the desired control goals.At present,the adaptive control of permanent magnet synchronous motor mainly has the following problems.On the one hand,the double closed loop of the permanent magnet synchronous motor adopts cascade control.The traditional PI controller has many parameters to be tuned;the tuning process is complicated;the speed loop and current loop PI controller parameters obtained by tuning have no obvious physical meaning.On the other hand,the parameters of permanent magnet synchronous motors are affected by factors such as temperature and magnetic saturation,which will reduce the dynamic performance of the system,and the control system is a strongly nonlinear,time-varying and multivariable system.High-performance control systems are essential.Therefore,the parameters needed for the motor control system need to be identified online.In view of the above two problems,it can be solved from the following aspects:(1)By using the particle swarm optimized Radial Basis Function(RBF)neural network algorithm to adjust the parameters of the PI controller,the number of parameters to be tuned by the PI controller is reduced,and the formula for PID parameter tuning by the RBF neural network is deduced.It facilitates the setting of the parameters of the speed loop controller of the permanent magnet synchronous motor,and realizes the adaptive control of the speed loop of the permanent magnet synchronous motor.(2)The parameter identification of motor inductance,resistance and permanent magnet flux linkage is realized by recursive least squares method with forgetting factor.The method has high identification accuracy,short consumption time,and good anti-interference ability,and its inductance identification error,flux identification error,and resistance identification error are relatively small.(3)By combining the recursive least squares parameter estimation with forgetting factor and the current loop PI controller,the parameter tuning process of the PI controller is simplified,and the current loop adaptive control of the permanent magnet synchronous motor is realized.The control quality and robustness of permanent magnet synchronous motors are improved.Finally,based on the neural network and the parameter identification algorithm,the hardware and software design of the permanent magnet synchronous motor is designed.The software and hardware implementation scheme of the entire control system is given.An experimental bench is built to verify the algorithm and analyze the experimental results.The experimental results prove the feasibility of the algorithm.