粗糙集神经网络在电机故障诊断中的应用  

Rough Set Neural Network in Fault Diagnosis of Motor

作　　者：康云霞[1] 尹作友[1] 

机构地区：[1]渤海大学,121000

出　　处：《电子测试》2013年第10期92-94,共3页Electronic Test

摘　　要：随着社会的进步,科学的发展,电力工程在日常生活中的重要性也日趋明显。在电力系统中,电机的稳定与否关乎到整个系统稳定。电机设备诊断中针对感应电机故障复杂、提取方法不足等问题,运用粗糙神经网络对变压器的故障进行诊断,通过分析电机单相瞬时功率,滤波后进行小波包分解,从得出故障特征变化率,并用以表征故障特征,以此作为电机故障的依据,运用粗糙集理论进行约简,将约简结果作为特征向量输入到RBF网络中。结果表明该方法诊断灵敏度高,可用于电机的故障诊断。Along with social progress and scientific development, power engineering importance in everyday life are becoming increasingly apparent. In the power System, the motor is stable or not related to the stability of the entire system. Electrical equipment for induction motor fault diagnosis of complex issues such as inadequate extraction methods, the use of rough neural network fault diagnosis for transformers, single-phase motor by analyzing the instantaneous power, filtered wavelet packet decomposition, fault characteristics derived from the rate of change, and used to characterize the fault feature, as a basis for motor failure, the use of rough set theory reduction, the reduction results as a feature vector input to the RBF network. The results show that the method diagnostic sensitivity, can be used for motor fault diagnosis.

关 键 词：粗糙集 神经网络 电机诊断 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TM307.1[自动化与计算机技术—控制科学与工程]