| 引用本文: | 张宇博,郝治国,林泽暄,杨松浩,刘志远,于晓军.基于深度字典学习的输电线路故障分类方法[J].电力自动化设备,2022,42(11): |
| ZHANG Yubo,HAO Zhiguo,LIN Zexuan,YANG Songhao,LIU Zhiyuan,YU Xiaojun.Transmission line fault classification method based on deep dictionary learning[J].Electric Power Automation Equipment,2022,42(11): |
|
| 摘要: |
| 针对当前输电线路故障分类识别方法存在的阈值整定复杂、人工智能算法可解释性不足等问题，提出了一种基于深度字典学习的输电线路故障分类方法。该方法利用稀疏性约束驱动字典自动提取样本中的故障特征，同时深度字典结构使得所提取的故障特征具有较好的层次性和物理含义，符合人对故障的直观认识，一定程度上解决了数据驱动型方法可解释性不足的问题。最后，通过PSCAD/EMTDC仿真验证了所提方法的有效性。 |
| 关键词: 输电线路 故障分类 稀疏表示 深度字典学习 |
| DOI：10.16081/j.epae.202204031 |
| 分类号:TM75 |
| 基金项目:国家自然科学基金资助项目(52007143)；中国博士后科学基金资助项目(2021M692526) |
|
| Transmission line fault classification method based on deep dictionary learning |
|
ZHANG Yubo1, HAO Zhiguo1, LIN Zexuan1, YANG Songhao1, LIU Zhiyuan2, YU Xiaojun2
|
|
1.School of Electrical Engineering, Xi’an Jiaotong University, Xi’an 710049, China;2.State Grid Ningxia Electric Power Co.,Ltd.,Yinchuan 750001, China
|
| Abstract: |
| In order to solve the problems of complex threshold setting and insufficient interpretability of artificial intelligence algorithms in current transmission line fault classification and recognition methods, a transmission line fault classification method based on deep dictionary learning is proposed. Driven by the constraint of sparsity, the fault features of the samples are automatically extracted. Meanwhile, the structure of a deep dictionary enables the extracted fault features to have definite hierarchy and physical meaning, which is consistent with human intuitive understanding. The proposed method solves the problem of insufficient interpretability of data-driven methods to a certain extent. Finally, the effectiveness of the proposed method is verified by the simulation in PSCAD/EMTDC. |
| Key words: transmission line fault classification sparse representation deep dictionary learning |
