基于SSA-PNN的电力变压器故障诊断

Fault diagnosis of power transformer based on SSA-PNN

为了提高变压器故障诊断的准确率,在改良三比值法的基础上,采用麻雀搜索算法优化概率神经网络构建一种新型变压器故障诊断网络模型,并设计相应的故障诊断方法。分析表明,与基于概率神经网络的变压器故障诊断方法相比,基于该网络模型的诊断方法提高了变压器故障识别与故障分类的准确率,在电力变压器的故障诊断中具有一定的实际工程意义。

In order to improve the accuracy of transformer fault diagnosis,based on the improved three ratio method,the sparrow search algorithm is used to optimize the probabilistic neural network,a new transformer fault diagnosis network model is established,and the corresponding fault diagnosis method is designed.The analysis shows that compared with the transformer fault diagnosis method based on probabilistic neural network,the diagnosis method based on the network model improves the accuracy of transformer fault identification and fault classification,and has certain engineering practical significance in the fault diagnosis of power transformer.