无刷直流电机故障诊断及容错控制技术

Fault diagnosis and fault tolerant control technology of brushless DC motor

针对无刷直流电机用霍尔位置传感器易故障问题,提出位置传感器故障诊断方法及容错控制技术。通过霍尔信号的Clark矢量变换,根据所得霍尔矢量相位的特点,提出了通过检测其相位变化诊断霍尔故障类型的方法。分析各种霍尔故障类型对转子位置估算精确度的影响,提出了两种容错控制策略:改进型混合观测器补偿法;改进型插值容错控制算法。在所搭建的电机控制试验平台上,通过模拟霍尔传感器故障实验,分别在稳态运行、动态运行(包括转速突变、变负载运行)时,验证所提算法的有效性和可靠性。实验结果表明,所提故障诊断法可快速识别霍尔故障类型,两种算法在单霍尔和双霍尔故障时都能准确地估计转子位置,有效补偿霍尔故障,而改进型插值容错控制算法具有更好的动态性能。

Aiming at the failure problem of Hall-effect position sensor for brushless DC motor,the fault diagnosis method and fault-tolerant control technology of brushless DC motor position sensor are proposed.A method,through the Clark vector transformation of the Hall signal and based on the characteristics of the obtained Hall vector phases,is proposed to diagnose Hall fault type by detecting their phases change.The influence of various faults on the accuracy of rotor position estimation was analyzed,and two fault-tolerant control strategies which are the improved hybrid observer compensation method and the improved interpolation fault-tolerant control algorithm were proposed.The effectiveness and reliability of the proposed algorithm was verified by simulating the failure of the Hall sensor in steady-state operation and dynamic operation(including acceleration transient and variable load)at the motor control test platform.Experimental results show that the proposed fault diagnosis method can quickly detect the type of Hall fault and the two algorithms accurately estimate rotor position in both single and double Hall faults to effectively compensate Hall fault.And the improved interpolation fault-tolerant control algorithm has better dynamic performance.