基于主成分分析和遗传程序设计的油浸式变压器故障诊断

提出了一种基于主成分分析和遗传程序设计的分类模型,并应用于油浸式变压器故障诊断中。该分类模型以变压器油中溶解气体的DGA数据为评价指标;通过对所采集数据进行主成分分析,减少了数据的冗余度,从而大大降低了问题的复杂度。实验证明该模型是高效可行的。

Oi-l immer sed t ransformer fault diagnosis based on PCA and GP

We proposed a classification model based on principal component analysis and genetic programming,and applied it to fault diagnosis of oil-immersed transformers.The classification model used the DGA data of dissolved gas in the transformer oil as the evaluation indicalors;It reduced data redundancy by using the PCA to analyse the collected data,which greatly reduces the complexity of the problem.Experimental results show that the model is efficient and feasible.