基于COMSOL的干式变压器故障诊断

Dry-type Transformer Fault Identification Based on COMSOL

干式变压器是一种重要的电力设备,广泛应用于电力系统和工业领域。传统的干式变压器故障检测方法主要依靠人工巡检和手动诊断,存在着准确性低、效率低、依赖性强等问题。随着机器学习技术的快速发展,基于机器学习的干式变压器故障识别方法逐渐得到了广泛应用。基于COMSOL Multiphysics软件的有限元分析方法,对干式变压器进行模型构建、网格划分和温度场仿真。使用仿真计算得到的数据进行机器学习训练集和测试集的制作。分析了机器学习多分类算法并对数据处理及分类准确率进行了比较分析。分析并研究了对支持向量机模型参数进行了遗传算法优化的故障识别方法,准确率可达96.00%,在对比传统支持向量机聚类方法上,准确率提高了11.63%。

Dry type transformer is an important power equipment,widely used in power system and industry.The traditional dry transformer fault detection methods mainly rely on manual inspection and manual diagnosis,which has some problems such as low accuracy,low efficiency and strong dependence.With the rapid development of machine learning technology,dry type transformer fault identification method based on machine learning has been widely used.In this paper,based on the finite element analysis method of COMSOL Multiphysics software,the model construction,mesh division and temperature field simulation of dry type transformer are carried out.The machine learning training set and test set are made by using the data obtained from the simulation calculation.Multi-classification algorithm of machine learning is analyzed and data processing and classification accuracy are compared.The fault identification method of SVM model parameters optimized by genetic algorithm is analyzed and studied.The accuracy rate can reach 96.00%.Compared with the traditional support vector machine clustering method,the accuracy is improved by 11.63%.