Research On PMSM Position Sensorless Control Strategy For Electric Vehicles

In recent years,with the popularization of new energy vehicles,especially electric vehicles with permanent magnet synchronous motors as the power source have been widely promoted.For the control system of a permanent magnet synchronous motor,the position information of the rotor is critical to the accuracy of the control.However,the traditional mechanical sensor not only increases the system cost,but also is affected by various complicated environmental factors and affects the further promotion of the permanent magnet synchronous motor.Therefore,the research on the position sensorless control algorithm of permanent magnet synchronous motor shows a high academic and application value.Firstly,a sliding mode variable structure algorithm based on adaptive adjustment of current error is used to estimate the rotor speed and position.The sliding mode variable structure algorithm can extract the back-EMF of the motor from the current and voltage of the stator,and calculate the rotation speed and position of the rotor.The proposed algorithm uses an adaptive function gain,and the traditional method cannot adjust according to the state of the system,which affects the estimation accuracy of the system and increases the jitter of the chatter.The proposed algorithm can adaptively adjust the gain of the sigmoidal function according to the magnitude of the current error,thereby reducing the chattering and error of the system and improving the performance of the system.The feasibility and superiority of the proposed algorithm are verified by simulation analysis under different speeds and loads.Secondly,based on the adaptive sliding mode variable structure,the PMSM position sensorless control system with phase-locked loop with feedforward compensation link is added.Since the back-EMF estimated by the sliding mode variable structure control has a high frequency component,which the direct numerical calculation of the rotational speed and position of the rotor will bring about chattering of the system.This chapter uses a new algorithm that combines sliding mode variable structure control with a phaselocked loop with feedforward compensation.Since the phase-locked loop has the characteristics of low-pass filtering,the high-frequency component contained in the backEMF can be filtered out,and the chattering of the system is removed,and the control precision and performance of the system are improved.Advance in the feedforward compensation link reduces the overshoot and instability before the system stabilizes.Finally,an AEKF algorithm at low speed is used to estimate the rotor speed and position.Since the PMSM' back-EMF is weak at low speeds,it is difficult to observe using a sliding mode observer.Therefore,this paper adopts an extended Kalman filter algorithm that adaptively adjusts the current data weight according to the system error covariance,which can take the stator current and voltage as input,the rotor speed and position as the output,and calculate the rotor position information online.And the proposed algorithm can adjust the weight of the current measurement data,weaken the historical data,and increase the estimation accuracy of the system.The effectiveness of the proposed algorithm is verified by analyzing the simulation data.