基于改进的神经网络在变压器故障诊断应用的研究

Transformer Fault Diagnosis Based on Neural Network

变压器在电力系统中有着重要作用,其安全性和可靠性是保证电网连续运行和供电的基础。针对变压器正常样本和故障样本类别分布不平衡的问题,对比验证传统的采样方法和生成对抗网络处理样本数据不平衡的问题,构造平衡样本数据集。仿真结果证明,数据集的平衡能有效提高神经网络对变压器故障诊断的正确率。

Transformer plays an important role in power system.Its security and reliability are the basis to ensure the continuous operation and power supply of power grid.Aiming at the problem of unbalanced distribution of transformer normal samples and fault samples,the balanced sample data set is constructed by comparing and verifying the traditional sampling methods and the problem of unbalanced processing sample data of the generated countermeasure network.The simulation results show that the balance of data sets can effectively improve the accuracy of neural network for transformer fault diagnosis.