<p>The sensorless control of surface-mounted permanent magnet synchronous motor (SPMSM) usually uses quadrature phase-locked loop (QPLL) to extract the phase information of the back electromotive force to realize the rotor angle estimation. However, the traditional QPLL has a convergence deviation of 180° when the motor is reversed, and the angle estimation error is obvious when the motor is accelerated and decelerated. To solve these problems, an enhanced QPLL (EQPLL) with polarity correction and high precision angle feedforward compensation is proposed. Firstly, the traditional phase discriminator is improved based on the two-phase stationary coordinate system, and the polarity correction function is designed by the error component of the improved phase discriminator to realize the non-convergent deviation angle estimation under the forward and reverse switching conditions of the motor. In addition, the error component of the improved phase discriminator is used as the feedforward compensation signal, and the enhanced generalized integrator is used to filter it, so as to realize the angle error compensation with low delay and low noise. Finally, the proposed scheme is verified by experiment on the motor platform, and compared with the existing scheme. The experimental results show that the proposed scheme can realize the polarity correction and angle error elimination, and at the same time, the noise mean square error is reduced by 24.33% compared with the existing angle feedforward compensation scheme.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Sensorless estimation of surface-mounted permanent magnet synchronous motors based on polar-corrected feedforward quadrature phase-locked loop

  • Kun Han,
  • Chen Yu,
  • Wei Li

摘要

The sensorless control of surface-mounted permanent magnet synchronous motor (SPMSM) usually uses quadrature phase-locked loop (QPLL) to extract the phase information of the back electromotive force to realize the rotor angle estimation. However, the traditional QPLL has a convergence deviation of 180° when the motor is reversed, and the angle estimation error is obvious when the motor is accelerated and decelerated. To solve these problems, an enhanced QPLL (EQPLL) with polarity correction and high precision angle feedforward compensation is proposed. Firstly, the traditional phase discriminator is improved based on the two-phase stationary coordinate system, and the polarity correction function is designed by the error component of the improved phase discriminator to realize the non-convergent deviation angle estimation under the forward and reverse switching conditions of the motor. In addition, the error component of the improved phase discriminator is used as the feedforward compensation signal, and the enhanced generalized integrator is used to filter it, so as to realize the angle error compensation with low delay and low noise. Finally, the proposed scheme is verified by experiment on the motor platform, and compared with the existing scheme. The experimental results show that the proposed scheme can realize the polarity correction and angle error elimination, and at the same time, the noise mean square error is reduced by 24.33% compared with the existing angle feedforward compensation scheme.