摘要
小型断路器作为使用最为广泛的开关电器之一,出现故障不仅会导致经济损失,严重时甚至会危害生命安全。因此,研究小型断路器的故障诊断具有非常重要的经济意义和工程应用价值。采用经验模态分解方法对小型断路器的分、合闸振动信号进行研究,提取其特征值,通过概率神经网络(PNN)进行故障诊断,并提供精确的仿真依据。仿真结果表明,PNN比BP神经网络在小样本的故障诊断中,辨识度更加精确,能够更有效地识别出小型断路器的故障。
The miniature circuit breaker is one of the most widely used switching appliances in daily life.The failure of the miniature circuit breaker will not only lead to serious economic loss,but also endanger life safety.It is of great economic and engineering significance to study the fault diagnosis of the miniature circuit breaker.The empirical mode decomposition method is used to study the opening and closing vibration signals of the miniature circuit breaker,and its characteristic values can be extracted.The fault diagnosis is carried out by the probabilistic neural network(PNN),which provides the accurate simulation basis for the fault diagnosis of the circuit breaker.Simulation results show that the PNN is more accurate than the BP neural network in the fault diagnosis of small samples,and can effectively identify the fault of the miniature circuit breaker.
作者
王兴宇
迟长春(指导)
张贤
WANG Xingyu;CHI Changchun;ZHANG Xian(School of Electrical Engineering,Shanghai Dianji University,Shanghai 201306,China)
出处
《上海电机学院学报》
2020年第4期222-227,共6页
Journal of Shanghai Dianji University
关键词
小型断路器
故障诊断
经验模态分解
振动信号
概率神经网络
miniature circuit breaker
fault diagnosis
empirical mode decomposition
vibration signal
probabilistic neural network(PNN)
