微特电机 ›› 2020, Vol. 48 ›› Issue (3): 43-47.

• 驱动控制 • 上一篇    下一篇

基于MRAS的异步电机无速度传感器应用研究

包广清, 祁武刚   

  1. 兰州理工大学 电气工程与信息工程学院,兰州 730050
  • 收稿日期:2019-07-08 出版日期:2020-03-28 发布日期:2020-03-30
  • 作者简介:包广清(1972—),女,博士,教授,博士生导师,主要研究方向为电机本体制造与控制。
  • 基金资助:
    甘肃省教育厅科技成果转化项目(2017D-05)

Application Research of Speed Sensorless Asynchronous Motor Based on MRAS

BAO Guang-qing, QI Wu-gang   

  1. College of Electrical Engineering and Information Engineering, Lanzhou University of Technology,Lanzhou 730050, China
  • Received:2019-07-08 Online:2020-03-28 Published:2020-03-30
  • About author:2020-3-43.pdf

摘要: 针对异步电机基于转子磁链模型参考自适应(MRAS)速度观测器中的积分直流偏置和PI自适应率参数不能在线修正的问题,将转子磁链的电压模型进行改进,并提出一种自适应线性神经元(ADALINE)速度观测器。通过一阶惯性环节代替了电压模型中的纯积分环节,消除了积分直流偏置和误差积累,利用ADALINE算法对MRAS观测器中的自适应率进行改进,实现了在中高速情况下对电机转速的准确观测。仿真结果表明,改进后的MRAS速度观测器对速度的辨识度较高,动态性能良好,验证了该控制策略的正确性与可行性。

关键词: 异步电机, 速度观测器, 无速度传感器, 自适应线性神经元, 模型参考自适应系统

Abstract: Aiming at the problem that the integral DC bias and PI adaptive rate parameters of speed observer based on rotor flux model reference adaptive (MRAS) can not be corrected online, the voltage model of rotor flux was improved and an adaptive linear neuron (ADALINE) speed observer was proposed. The first-order inertia link replaced the pure integral link in the voltage model, eliminating the DC bias and error accumulation. The ADALINE algorithm was used to improve the adaptive rate of MRAS observer and realize the accurate observation of motor speed at medium and high speed. The simulation results showed that the improved MRAS speed observer has high speed identification and good dynamic performance, which verifies the correctness and feasibility of the control strategy.

Key words: asynchronous motor, speed observer, speed sensorless, adaptive linear neuron (ADALINE), model reference adaptive system (MRAS)

中图分类号: