Research On The Sensorless Control Of Permanent Magnet Linear Synchronous Motor Based On High-Frequency Signal Injection Method

In order to achieve high-performance control of permanent magnet linear synchronous motors(PMLSM),vector control is used in the motor control systems.The position sensors can help to obtain the magnetic pole position required for vector control,but it will increase the cost and the environmental sensitivity of the control system.Therefore,position sensorless control has attracted the attention of scholars from various countries.At low speed,the high-frequency signal injection method based on the motor saliency is used in the position estimation system.In order to extract and separate the signal during the position estimation process,the FIR filters are needed,which will affect the dynamic performance.Therefore,this paper optimizes the FIR filter design method to improve the performance of the position estimation system.First,the mathematical model of the PMLSM and the realization method of vector control are given.Considering the differences between different modules,a multi-step simulation model is built in MATLAB/Simulink,which provides the foundation for the subsequent optimization research of the high-frequency signal injection method.After analyzing and summarizing the relevant literature,this paper introduces the mathematical model of the high-frequency signal method and its different implementation forms.The structure diagram of the position estimation system is also given.Based on the properties of the second-order system,the influence of the time delay is analyzed,which clarifies the significance of reducing the filter order.Then,in order to reduce the delay time caused by the FIR filter,this paper improves the two filtering parts required in the position estimation process.For the first-stage filtering,the combination of differential operation and low-pass FIR filter is used to replace the band-pass FIR filter.For the second-stage filtering,based on the analysis of the motor current components,the amplitude-frequency characteristics at some special frequencies are mainly considered.By using this method,an FIR filter with an order of only 3 is designed for position error signal extraction.In order to achieve simulation verification,a magnetic pole position estimation module is added in the vector control system.By comparing the position estimation results with different filter scheme,the effectiveness of the new filter scheme in this paper is verified.Finally,this paper introduces the design method of the software and hardware used in the sensorless control system.The model-based design method is used to shorten the time spent on software development.In addition,through the initial position estimation experiment,the effectiveness of the new filter scheme based on low-order FIR filter is further verified.