Fault Analysis And Fault Tolerant Control On Position Sensor Of Permanent Magnet Synchronous Motor For New Energy Vehicle

In order to cope with the oil crisis and environmental pollution,countries all over the world are doing their utmost to promote new energy vehicles.The permanent magnet synchronous motor vector control system is widely used in new energy vehicles,and the control accuracy of the system depends on the stability and accuracy of the feedback,especially the precision of the position sensor.However,the position sensor will inevitably fail due to the complexity of the working environment,which will affect the feedback,cause the vehicle to lose control and even threaten personal safety.Based on the situation,a new control method position sensorless was designed to be suitable for the failure,the fault diagnosis scheme and transition solution from position sensor control to the position sensorless control were presented in this thesis.With the new design,if the position sensor fails suddenly while the vehicle is running,the control system can automatically and smoothly transit to the position sensorless to ensure system stability.Firstly,the interior permanent magnet synchronous motor(IPMSM)vector control system is a nonlinear and strongly coupled system.In order to improve the anti-interference ability and response speed of the system,a sliding mode controller was used to replace the traditional PI controller.Secondly,aiming at the problem of complex and limited demodulation of resolver under cosine excitation,the square wave excitation and the corresponding demodulation strategy were used to improve the accuracy of resolver demodulation,and phase-locked loop tracking algorithm was used for soft decoding.Finally,based on permanent magnet synchronous motor position sensor fault,the possible failure types of the resolver were analyzed,and two kinds of fault diagnosis were designed to detect the failure in time and accurately.In order to solve the excessive oscillation problem caused by direct switching when the failure occurs.Two control strategy switching algorithms were optimized,and the curve weighting function switching algorithm was proposed to replace the direct switching and linear weighting function switching algorithms to ensure the smoothness and stability of the transition period.A fault-tolerant scheme based on position sensorless vector control was proposed.A sliding mode observer based on equivalent feedback was designed.Meanwhile,a piecewise power function was used to replace the sign function of the ideal sliding mode.Compared with the traditional sliding mode observer algorithm,it reduces the estimation error at low speed and weakens the chattering phenomenon in the estimation method.The model was built by Matlab / Simulink software,and the soft decoding algorithm,fault diagnosis,switching algorithm and improved sliding mode observer after fault of the rotating transformer were simulated and verified.The results show that the error of rotor position angle and speed obtained by the resolver soft decoding algorithm is small but the accuracy is high.The fault diagnosis scheme can determine the fault in time,and the curve weighting function switching strategy can make the transition smoother;Improved sliding mode observer estimates the rotation and position angle more accurately,the estimated speed chattering phenomenon is weakened,and the low speed error is reduced.It is verified that the overall scheme can realize the fault analysis and fault-tolerant control of the motor position sensor,the motor can still run normally even after the fault of position sensor,so as to ensure the safety of personnel and vehicles.