为提高动态绕组匝间短路故障的检测能力,提出了一种新的同步电机转子绕组匝间短路早期故障检测方法,通过分析同步电机转子数据,结合灰色关联度和主成分分析方法,构建了蚁狮算法与支持向量机的模型,提取关键故障数据作为支持向量机模型的输入变量,使用改进的蚁狮算法来优化支持向量机算法的关键参数,通过故障数据验证故障诊断模型.结果表明,基于XA-LO-SVM的故障诊断模型诊断精度可达97%以上,同时也缩短了诊断时间.
This paper aims to improve the detectability of interturn short circuit fault in dynamic winding and proposes a new method for early fault detection of interturn short circuit in rotor winding of synchronous motor.By analyzing rotor data of synchronous motor,combined with grey correlation degree and principal component analysis method,the approach works by developing a model of antlion algorithm and support vector machine;exacting the key fault data as input variables of the support vector machine model;optimizing the key parameters of SVM algorithm by using the improved antlion algorithm;and ver-ifying the fault diagnosis mode by fault data.The results show that the diagnosis accuracy of the fault di-agnosis model based on XALO-SVM can reach over 97%,and the diagnosis time is shortened,which provides technical support for fault detection.