基于引入欧拉算法的灰色预测理论的双绕组容错永磁电机驱动系统开路故障诊断策略研究

Dual-winding fault-tolerant permanent magnet motors (DFPMM) have the characteristics of high-power density, high efficiency, stable operation, and strong fault-tolerant ability. They are effectively used in military and civilian electric drive systems. Electrical faults objectively exist, and any single electrical fault will affect the reliability of the electric drive system. Fault diagnosis and fault-tolerant operation are key technologies for stable operation of motor drive systems. Aiming at the problems of traditional open circuit fault (OCF) diagnosis methods that are not fast enough, not intelligent enough, and prone to misdiagnosis when the load suddenly changes, a fault diagnosis strategy for DFPMM drive systems based on the introduction of Euler algorithm's grey prediction theory is proposed. The grey prediction theory can use less data to estimate and predict the changes in the system by using a grey prediction model that introduces the Euler algorithm. It has the advantages of requiring less sample data, high prediction accuracy, and simple operation. The original GM (1,1) model is optimized. By changing the smoothness of the original data sequence, the Euler algorithm is introduced to iteratively solve the background value and optimize the time response function in the prediction model to improve the accuracy of the traditional grey prediction GM (1,1) model. Through theoretical analysis, simulation, and experimental results, it has been verified that this strategy can accurately diagnose and locate the open circuit fault of the winding power transistor in the system in real time, improving the reliability of the electric drive system.