Sliding Mode Observer Based Position Sensorless Ipmsm Control

Position sensorless interior permanent magnet synchronous motor (IPMSM) control techniques have become a hotspot, which can reduce the cost and increase the reliability of drive system. In this thesis, the emphases lie on the sliding mode observer (SMO) applied for high-speed region, while the high-frequency (HF) voltage injection method that fits for standstill and low-speed region is used for rotor initial position detection and motor start. These two speed regions are smoothly linked up by a proper switching principle.Firstly, the HF mathematical model of IPMSM is presented. The HF current excited by the HF fluctuating voltage injection is demodulated to extract the rotor magnetic pole position. According to the magnetization curve of the core, the pulse voltage vector injection is adopted to identify the magnet polarity by comparing the peak value of the two excited currents, from which the rotor angle and speed at standstill and low-speed region can be gained.Secondly, on the basis of fundamental frequency mathematical model of IPMSM, an extended electromotive force (EMF) based SMO is constructed based on the sliding mode variable structure. The traditional sgn-function is replaced by sat-function to alleviate the chattering problem of the sliding mode operation. In addition, a self-adaption low-pass filter is implemented to make the phase delay a constant value at different rotor speeds by adjusting its corner frequency in real time, which can accurately acquire the rotor angle and speed at high-speed region and efficiently improve the dynamic estimation performance.Finally, a hardware and software system for position sensorless IPMSM vector control is designed, on which the experimental verifications are carried out. The experimental results show that:the HF voltage injection method can identify the rotor pole angle accurately, which enables the IPMSM operate stably at standstill and low-speed region; the proposed SMO with self-adaption low-pass filter can improve the dynamic estimation performance, which enables the IPMSM operate stably at high-speed region; the switching principle can guarantee smooth transition between low-speed and high-speed regions. The vector control of the position sensorless IPMSM in the whole speed range is verified.