| 李文悦,何怡刚,邢致恺,周亚中,雷蕾潇.基于双输入残差图卷积网络的电力变压器健康状态评估方法[J].电子测量与仪器学报,2024,38(11):15-24 |
| 基于双输入残差图卷积网络的电力变压器健康状态评估方法 |
| Health evaluation of power transformer based on doubleinput residual graph convolutional network |
| |
| DOI： |
| 中文关键词: 电力变压器 状态评估 图卷积网络 图池化 数据不平衡 |
| 英文关键词:power transformer health evaluation graph convolutional network graph pooling data imbalance |
| 基金项目:国家重点研发计划“储能与智能电网技术”专项“海上风电并网系统远程监测与故障诊断技术”项目（2023YFB2406900）资助 |
|
|
| Author | Institution |
| Li Wenyue | School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072，China |
| He Yigang | School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072，China |
| Xing Zhikai | School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072，China |
| Zhou Yazhong | School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072，China |
| Lei Leixiao | School of Electrical Engineering and Automation, Wuhan University, Wuhan 430072，China |
|
| 摘要点击次数: 468 |
| 全文下载次数: 1266 |
| 中文摘要: |
| 电力变压器运行过程中故障数据远少于正常数据，数据不平衡问题较为严重，并且监测变量间耦合关系复杂，导致状态评估任务的建模难度大、评估精度低。针对相关问题，提出基于双输入残差图卷积网络的电力变压器健康状态评估方法。首先，采用SMOTE Tomek混合采样算法对训练数据进行不平衡数据预处理，解决了故障数据过少、训练数据严重不平衡的问题；然后，考虑到单一度量无法准确描述变量间相关性的问题，提出多度量融合构图方法，通过多个度量方法共同学习变量间的相关性，并构造图结构数据；最后，提出基于切比雪夫图卷积的双输入残差图卷积网络，对所构造的图结构数据进行特征提取，并通过自注意力机制进行特征融合，得到变压器的状态评估结果。在真实电力变压器运行的油中溶解气体及油化试验数据集上进行了对比实验，实验结果表明，所提出方法的状态评估准确率达到94.7%，F1分数达到0.942，高于其他深度学习方法，具有最佳的评估性能。 |
| 英文摘要: |
| The operation of power transformers involves significantly fewer fault data compared to normal data, resulting in a severe data imbalance issue. Additionally, the complex coupling relationships among the monitored variables make the modeling of condition assessment tasks challenging and lead to low evaluation accuracy. Aiming at the related problems, a power transformer health condition evaluation method based on double input residual graph convolutional network is proposed. First, SMOTE Tomek mixed sampling algorithm was used to pre-process the unbalanced data of the training data, which solved the problem of insufficient fault data and difficult classification. Then, a multi-metric fusion graph construction method is proposed to learn the correlation between variables from multiple variables and construct the graph structure data. Finally, a double input residual graph convolutional network(DI-ResGCN) based on the ChebyNet is proposed, feature extraction is carried out on the constructed graph structure data, and feature fusion is carried out through the self attention mechanism to obtain the transformer health evaluation results. Experiments were carried out on a dataset of dissolved gases in oil and oil test collected by a real power transformer, and experimental results show that the proposed method has a state assessment accuracy of 94.7% and F1 score of 0.942, outperforming other deep learning methods and exhibiting the best evaluation performance. |
| 查看全文 查看/发表评论 下载PDF阅读器 |
|
|
|
