基于多项式混沌观测器的电机系统自愈控制

Self-healing Control for Motor System Based on Polynomial Chaos Observer

针对电机控制系统中电流传感器故障提出了一种自愈控制方式。考虑转子电阻不确定性,结合多项式混沌理论与卡尔曼滤波算法,设计了多项式混沌电流观测器,基于该观测器实现对电机系统中电流传感器的实时监测、故障诊断和故障隔离。在传感器正常时,利用观测器观测值进行故障诊断;当故障发生时,隔离传感器,观测器反馈观测值给控制电路,从而实现对系统的自愈控制,提高了系统稳定性。仿真结果验证了该方法的有效性。

This paper proposes a self-healing control method for sensor faults in the motor drive system. Considering the uncertainty of rotor resistance in the system, a polynomial chaos observer is designed by combining polynomial chaos theory with Kalman filter. With the observer, real-time monitoring, fault diagnosis and isolation of the current sensor are fulfilled. When sensors are in healthy condition, observed values of the observer are utilized to fault diagnosis; if faults occur in current sensor, the sensor is isolated, the observer feeds back observed values to the control circuit, thereby the goal of self-healing control for the motor system is achieved. The method improves the stability of the system. Finally, simulation results demonstrate the effectiveness of the method.