基于互补免疫算法的变压器故障诊断

Fault Diagnosis of Transformers Based on Complementary Immune Algorithm

基于自组织抗体网络(so Ab Net)的变压器故障诊断方法中没有网络压缩机制,并且网络的初始抗体是随机选取的,网络性能不稳定。针对这一问题,提出了基于互补免疫算法的变压器故障诊断方法,结合变压器故障诊断的特点详细设计了免疫算子以弥补so Ab Net的不足。免疫算子中接种疫苗利用K-means最佳聚类算法为so Ab Net提供初始抗体,并通过免疫选择压缩网络规模,其参数由粒子群算法进行优化。变压器故障诊断实验结果表明,所提出的互补免疫算法能够充分利用系统的先验知识,并有效地提取故障样本的数据特征,与单一智能方法相比具有更高的诊断准确率。

The transformer fault diagnosis based on self-organization antibody net（so Ab Net） has no network compression mechanism and selects the initial antibodies randomly, so its network performance is instable. Thus, a diagnosis method based on complementary immune algorithm for power transformer is proposed in this paper, and immune operator is designed in detail considering the characteristics of transformer fault diagnosis. Vaccination of immune operator uses K-means optimal clustering algorithm to provide initial antibodies for so Ab Net and compresses the network through immune selection, and its parameter is optimized by particle swarm optimization（PSO） algorithm. Experimental results demonstrate that the proposed complementary immune algorithm could make use of prior knowledge and extract the data characteristics of the fault samples effectively. The diagnostic accuracy of the proposed algorithm is higher than that of the single intelligence algorithm.