Research On Online Parameter Identification And Sensorless Control Of Permanent Magnet Motors Considering Inverter Nonlinearity

Due to the complexity of the vehicle operating environment,the wheel hub motors and their subsystems in intelligent wire-controlled vehicle chassis operate under non-steady-state conditions such as frequent start-stop events.The high integration level leads to increased motor temperature rise,which can cause variations in parameters of permanent magnet synchronous motors(PMSMs),such as magnetic flux and resistance.Currently,real-time and accurate rotor position acquisition is essential for efficient motor control.However,traditional position detection devices are bulky and unsuitable for highly integrated systems.Moreover,operating in harsh environments can lead to sensor failures,impacting motor reliability.Therefore,studying the efficient,safe,and reliable control of the executing motors in the line-controlled chassis,including online parameter identification of PMSMs and the application of sensorless control algorithms,is of significant research importance for optimizing chassis control,fault diagnosis,and ensuring safe and stable vehicle operation.This paper focuses on investigating the online parameter identification and sensorless control methods for the executing motors in the line-controlled chassis to achieve efficient,stable,and redundant control objectives.Firstly,an overview of the research background and current status of permanent magnet synchronous motors is provided,leading to the research objectives.The ideal mathematical model of PMSMs,coordinate transformation forms,and observability analysis of the model are described in detail,and an accurate vector control simulation model is constructed.Secondly,to address the issue of parameter identification accuracy caused by the nonlinearity of inverters,a full-parameter identification technique for surface-mounted permanent magnet synchronous motors(SPMSMs)based on the dual extended Kalman observer is proposed.With this technique,a dual extended Kalman filter-based multi-parameter observer considering the nonlinearity of inverters is constructed for the multi-parameter identification of PMSMs and deadzone compensation of inverters.Simulation and experimental results show that the observer can effectively compensate for the influence of voltage and current disturbances caused by dead-zones on parameter identification within the global operating range of the motor,resulting in a multiparameter identification error of less than 5%.The observer exhibits excellent parameter identification performance and algorithm robustness with insensitivity to initial parameter errors.Thirdly,an interactive multiple model extended Kalman filter-based sensorless control algorithm is proposed in this study.By optimizing observer design,selecting appropriate models,and considering the disturbance voltage of the inverter and parameter identification simultaneously under the premise of observability,sensorless control with improved speed estimation accuracy and a wider range of applicability is achieved.The effectiveness and robustness of the proposed algorithm are verified through simulations and experiments,demonstrating good adaptability to wide speed ranges and varying loads.Finally,to validate the effectiveness and disturbance rejection capability of the parameter identification and sensorless control algorithms in practical environments,a single-motor and a dual-motor towing test bench are constructed.The hardware and software designs of the two test benches are described in detail,and the stability and anti-interference capability of the proposed algorithms are validated through the design of various operating conditions.