张镱议,彭鸿博,李昕,赵刘亮,郑含博,刘捷丰.基于DGA特征量优选与改进磷虾群算法优化支持向量机的变压器故障诊断模型[J].电测与仪表,2019,56(21):110-116. Zhang Yiyi,Peng Hongbo,Li Xin,Zhou Liuliang,Zheng Hanbo,Liu Jiefeng.Fault Diagnosis Model of Power Transformers Based on DGA Features Selection and Genetic Algorithm Optimization SVM[J].Electrical Measurement & Instrumentation,2019,56(21):110-116.
基于DGA特征量优选与改进磷虾群算法优化支持向量机的变压器故障诊断模型
Fault Diagnosis Model of Power Transformers Based on DGA Features Selection and Genetic Algorithm Optimization SVM
In order to make up a single feature gas or feature ratios as DGA feature can not fully reflect the transformer fault, a preferred DGA feature set generate by the hybrid DGA feature set are used as the input vectors, and a transformer fault diagnosis model based on the improved krill herd (IKH) optimization support vector machine (SVM) is proposed. Firstly, the SVM parameters and 11 features are encoded by a binary code technique. Then, a preferred DGA feature set for fault diagnosis of power transformers is selected by Genetic Algorithm (GA) and SVM. Finally, optimizing the SVM parameters by GA, and conducting a transformer fault diagnosis model based on IKH algorithm optimization SVM combined with the cross validation principle. The fault diagnosis results based on IEC TC 10 show that the proposed DGA feature set increase the accuracy by 10.14% and 30.2% over the DGA data and IEC ratios. Furthermore, the accuracy of IKHSVM is better than the standard SVM and GASVM (accuracy rate are 73.87%、81.13% and 86.27%) that verify the accuracy of transformer fault diagnosis can be improved by the proposed method.
