摘要
利用变压器的油色谱试验数据对神经网络进行训练,将训练后的神经网络用于变压器的状态诊断。同时利用遗传算法优化神经网络的结构。算例结果表明,利用遗传算法和神经网络相结合的人工智能方法可以有效地诊断变压器的状态。
This paper presents a method to diagnose the state of transformer based on artificial intelligence which integrates artificial neural network and genetic algorithm. In this method, ANN is used as a classifier to forecast the state of transformer after training the samples of oil-chromatogram examination, and genetic algorithm is used to optimize the ANN structure. The practical result shows the validity of the method.
出处
《高压电器》
CAS
CSCD
北大核心
2004年第3期228-230,共3页
High Voltage Apparatus
关键词
变压器
状态检修
人工神经网络
遗传算法
transformer
condition based maintenance
artificial neural network(ANN)
genetic algorithm
