基于支持向量机及油中溶解气体分析的大型电力变压器故障诊断模型研究

提出用支持向量机作为分层决策电力变压器故障诊断模型。首先通过相关统计分析,选择典型油中气体作为支持向量机输入参数,然后在深入发掘油中气体所含故障信息基础上,利用典型故障气体的相对含量在高维空间的分布特性进行变压器故障类型诊断。该方法基于小训练样本条件下寻求最优解,具有很好的推广能力及一致性等优点,还适用 于变压器典型故障数据少的特点。文中还给出了两种不同支持向量机核函数分类结果的比较。为了提高故障诊断的正判率,该模型同时在相关性强的特征气体之间,利用K-近邻搜索聚类在最优分类面附近对分类结果进行精确逼近,使分层决策模型可靠性显著改善。计算结果表明,该模型具有很好的分类效果。

FAULT DIAGNOSIS MODEL FOR POWER TRANSFORMER BASED ON SUPPORT VECTOR MACHINE AND DISSOLVED GAS ANALYSIS

A multi-level decision-making model for power transformer fault diagnosis based on SVM (Support Vector Machine) is presented. Based on correlation analysis, some key gases are selected as the inputs of SVM; furthermore, improving the use of the fault information within DGA (Dissolved Gas Analysis), the fault diagnosis is accomplished according to the concentration distribution of typical fault gases in higher dimensional space. The proposed approach is based on seeking the optimal solution by few training samples supporting, and it has important features such as good generalization and consistency performance, etc, which is very suitable to solve the problems of less typical fault data for diagnosis. And the comparison between two kinds of SVM is presented also. Meanwhile the output of this model is improved by approaching exactly with K-Nearest Neighbor Search Classification for the SVM classification results, which is adjacent to optimal separating hyperplane. So the dependability of this model is enhanced greatly, and its effectiveness and usefulness is proved.