摘要
监测可能发生故障的电力电子器件,对电力电子装置的故障进行识别和诊断,以降低电气系统的故障发生率,对于减少风力发电机组运行的故障率,降低风力发电运行维护成本有着重要意义。为此,提出将自组织特征映射神经网络(SOM)应用于风力发电机组电力电子装置的故障诊断中。实验结果表明,利用该方法进行风力发电机电力电子装置故障诊断能取得较好的效果,具有一定的工程应用价值。
Monitoring possible failures of power electronic devices could identify and diagnose faults of power electronic devices,which can not only reduce the incidence of electrical system failures and the failure rate of the wind turbine, but also reduce operation and maintenance costs of wind power generation. In this paper, a self-or- ganizing feature map（SOM） neural network was applied to fault diagnosis of wind turbine power electronic devices. Experimental results show that the proposed method can achieve better results and has a certain value of engineering application.
出处
《电力系统及其自动化学报》
CSCD
北大核心
2010年第3期142-145,共4页
Proceedings of the CSU-EPSA
关键词
风力发电机组
电力电子装置
SOM神经网络
wind turbine
power electronic devices
self-organizing feature map（SOM） neural network
