基于神经网络的无刷直流电机故障诊断和容错控制方法的研究

提出并设计了一种针对无刷直流电机控制器的故障诊断和容错控制方法。该神经网络状态观测器由实际非线性控制系统进行训练,根据实际系统和神经网络观测器输出残差,容错控制系统能够被确定下来。针对系统控制器、电流和速度传感器的故障条件进行了仿真分析,容错控制通过使用补偿控制器实现,保证系统故障条件下的稳定性和性能。结果表明,所提出的方法能有效地定位错误故障和时间,消除干扰并提高系统的鲁棒性。

Study on fault diagnosis and fault tolerant control of BLDC motor based on neural networks

A fault diagnosi and fault tolerant control system for brushless direct current motor controller was designed.The neural network state observer is trained by real nonlinear control system.Based on the residual difference between outputs of actual system and neural network observer,the fault of control system is detected and determined.Simulation of system controller and current and speed sensors faults is implemented.Fault tolerant control is realized by using compensation controller which can guarantee the...