Research On Position Sensorless Control Of Permanent Magnet Linear Synchronous Motor

Due to the continuous development of industrial technology, the traditional linear feeding system cannot fulfill the requirement of the precise processing technique and the high-speed evolution in the instrumentation industry.The feeding system of permanent magnet linear synchronous motor(PMLSM) was widely applied in assembly manufacture and machining industries because of its high speed, accurate positioning and large motion distance. However, the traditional PMLSM feeding system with physical sensor still has many defects. To solve the issues caused by importing the mechanical sensor, developing the position sensorless control system with positive and economic servo performance becomes a hot research topic in current years. This thesis is aimed at solving the core problems on the PMLSM: how to estimate the real-time rotor position and speed.Firstly, the operating principle and composition structure of the PMLSM has been researched in detail in this thesis. Then the sensorless position control project for PMLSM is constituted based on analysis of the theory of SVPWM and the traditional feeding system. And the feedback signals of position and speed are got from estimator.Secondly, the Augmented Extended Kalman Filter(AEKF) is presented in this thesis to estimate precisely the rotor position and speed during full speed range. This estimation algorithm can estimate the system state and winding resistances at the same time, avoiding the inaccurate estimation problem causing by winding resistances change. In the other part, in order to obtain the initial position of magnet polarity to make sure the motor can start properly, injecting voltage pulse vectors method has been proposed in this thesis, and the estimation accuracy in theory is??9375.0. A kind of incremental Fuzzy-PID controller is added in the speed loop to improve the robustness and disturbance of the whole system.Lastly, the software and hardware for PMLSM are built based on simulation.The estimated project for sensorless servo system of PMLSM is verified on Gantrydual-drive linear motor platform. As the experimental results suggest that the estimated value of displacement and velocity can match well with the output of grating measure, and the estimated error is in a reasonable range. Above all, this project can satisfy the most linear motor servo system.