改进卷积神经网络在互感器故障诊断中的应用

Application of Improved Convolution Neural Network in Fault Diagnosis of Transformer

低压电流互感器作为电网中的关键设备,已经得到广泛使用。低压电流互感器故障诊断的在线检定也显得十分重要。提出了一种改进的全局平均池化的一维卷积神经网络(1DCNN-SVM)故障诊断模型应用于低压电流互感器在线检定。该方法改进了传统卷积神经网络(CNN)模型的结构,引入全局平均池化而不是全连接网络结构,并在测试阶段使用支持向量机(SVM)替代Softmax函数。通过进行实验分析,将所提的方法与传统的CNN进行实验对比,实验结果表明所提方法在训练时间、测试时间以及模型的测试精度等方面的表现都比传统的CNN结构模型要好。

Low-voltage current transformer is a key equipment in the power system.In the on-line verification of lowvoltage current transformer,it is very important to classify and identify fault samples.In this paper,an improved global average pooling 1DCNN-SVM fault diagnosis model is proposed for on-line verification of low voltage current transformer.The proposed method improves the structure of the traditional CNN model,introduces global average pooling to replace the fully connected network structure,and uses SVM to replace Softmax function in the test stage.Through experiments,the proposed method is compared with the traditional CNN.The experimental results show that the proposed method is better than the traditional CNN structure model in terms of training time,testing time and test accuracy of the model.