Research On Sensorless Control Strategy Of Surface Mount Double Rotor Permanent Magnet Synchronous Motor

The counter-rotating propeller system driven by the counter-rotating motor has the advantages of high energy efficiency and no roll supercharging,so it is widely used in the fields of wind power generation,aircraft and underwater propulsion.The control system of the counter-rotating permanent magnet synchronous motor requires two position sensors to complete the vector control.The two position sensors exceed the volume of the larger system,which increases the cost and reduces the overall reliability.Sensorless control strategy avoids these above problems.Therefore,this paper studies the vector control system and position sensorless control strategy of dual rotor surface-mounted permanent magnet synchronous motor(DR-SPMSM).The main work is as follows:First,according to the mathematical model of the dual-rotor motor,its vector control system is studied.The DR-SPMSM can only control the relative speed of the two rotors,and cannot control the speed of each rotor separately.Therefore,the counter-rotating propeller system driven by the dual-rotor motor is analyzed.In order to improve the dynamic response of the motor,an improved load torque observer is proposed.On this basis,torque feedforward compensation control is implemented.Finally,the foregoing contents are verified through simulation.Secondly,the sliding mode observer is used to study the sensorless control algorithm of the motor at high speed.UNder the discrete sliding mode control theory,the reason for the chattering of the traditional sliding mode observer is analyzed.An improved discrete sliding mode observer is proposed to reduce chattering.A discrete sliding mode observer is designed based on the combined approach law,and the estimated back EMF is introduced as a feedback term.The adaptive back-EMF filter is used instead of the low-pass filter to obtain the estimated back-EMF.The steady-state error generated by the phase-locked loop during the acceleration and deceleration of the motor is studied and compensated.Finally,the foregoing contents are verified by simulation..Thirdly,for the problem that the motor model method obtains the rotor position information sensitive to the motor parameters,the online paramete r identification of the motor under the control of the position sensor is studied.The changes of various parameters of the motor during the operation of the motor and its influence on the model method are analyzed.The observability of the motor under dif ferent control strategies is analyzed through the observability criterion.A step-by-step identification strategy is proposed,and an online identification of two parameters is realized by extended Kalman filter.Finally,the foregoing contents are verified by simulation..Finally,the sensorless control algorithm of the motor at low speed is studied.On the basis of the saturation convex polarity of the surface-mounted motor,the method of high frequency pulsed square wave signal injection is studied.The separation method of fundamental frequency signal and high frequency signal based on moving average filter is studied.Aiming at the wide-speed operation of the motor,the switching strategy of the compound control algorithm is studied.Finally,the foregoing contents are verified by simulation.