Sensorless Control Of Permanent Magnet Synchronous Motor With Online Parameter Identification

Permanent magnet synchronous motors(PMSM)have been widely used in the fields of electric vehicles,rail transit,agriculture equipment,ships and household appliances.The control of PMSM needs to obtain the rotor position.Position sensor is complicated and increases the cost of the system.Therefore,the method of sensorless control has been brought into focus.However,during the operation of PMSM,the motor parameters will change under different working conditions and affect the rotor position acquisition,thus affecting the efficiency of the control system.Therefore,it is of great significance to study the sensorless control of PMSM with online parameter identification.First,mathematical modeling of PMSM in three common coordinate systems is carried out.several common control strategies are discussed.A double closed loop control system of PMSM current-speed is established.The speed loop and current loop controller parameters are set,and the correctness and reliability of PMSM mathematical modeling and control system are verified by simulation.Secondly,the theory of sliding mode are studied.A sensorless control system based on SMO is designed and verified by simulation.In this paper,the influence of motor parameters changes and the observability of motor parameters on sensorless control system are studied,and an implicit pole PMSM online parameter identification algorithm based on Recursive least Square(RLS)method is proposed,and the influence of the selection of variables and fitted equation on identification results is analyzed.This method identifies the quadrature axis inductance and feeds the results back to sliding mode observer while updating the current controller parameters so that the online parameter identification and real-time update of controller parameters are realized,effectively improves the accuracy of rotor position detection and the operating performance of the system.In order to compare the performance of RLS single-parameter identification,the quantum particle swarm(Quantum-behaved PSO,QPSO)algorithm was introduced into the parameter identification of PMSM model,and a PMSM multi-parameter identification model based on QPSO algorithm was proposed,and the identification results of the two algorithms were simulated and analyzed.Finally,a PMSM experiment platform based on RLS is designed and built.Hardware circuits contains main circuit,current sampling circuit,DAC circuit.The software part mainly includes sliding mode observer,RLS online parameter identification algorithm,speed-current loop.The influence of PMSM parameters on sensorless control system is verified by experiments.The performance of sensorless control system with&without online parameter identification is compared.Experiment verified the rationality and feasibility of the method brought by paper.