基于因子分析和基因表达式编程算法的变压器故障诊断

将因子分析和基因表达式编程结合起来,对变压器油色谱数据进行处理和智能训练,并建立变压器故障分类诊断模型;收集到332组油色谱数据,选取150组作为GEP分类器的训练样本,182组作为测试样本,并将测试结果与三比值法和朴素贝叶斯分类器进行比较。大量诊断实例表明,所提出的变压器故障诊断模型性能优于另外两种方法。

Transformer fault diagnosis based on factor analysis and gene expression programming

The factor analysis and gene expression programming are combined to go on with the transformer oil chromatography data processing and intelligence training,and the transformer fault diagnosis model classification is establish.332 groups of oil chromatographic data are collected,and 150 groups are selected as the training samples of GEP classifier,100 groups as the test samples.Test result is compared with those of three-ratio method and the naive Bayesian classifier.A large number of diagnostic examples show that the proposed transformer fault diagnosis model is superior to the other two methods.