| 引用本文: | 郭倩雯,莫文雄,郑方晴.一种基于部分Hausdorff距离的励磁涌流识别新方法[J].电力系统保护与控制,2019,47(1):35-42.[点击复制] |
| GUO Qianwen,MO Wenxiong,ZHENG Fangqing.A new method of inrush current identification based on partial Hausdorff distance[J].Power System Protection and Control,2019,47(1):35-42[点击复制] |
|
| 摘要: |
| 变电站的电磁环境较为复杂，二次设备可能受到电磁干扰等不利因素影响而产生异常数据，进而导致保护误动。传统的变压器励磁涌流识别方法并没有考虑到异常数据干扰对保护的影响。分析异常数据对不同识别方法造成的影响，可以得知现有的励磁涌流识别方法在保护用数据受到异常干扰时，可能会导致保护动作速度变慢甚至保护误动。为解决这一问题，提出了一种基于波形相似度识别的励磁涌流快速识别方法，通过对比差动电流和正弦信号的波形相似度以区分励磁涌流和故障电流。仿真结果表明，该方法能够快速有效识别励磁涌流，在一定程度上能够不受异常数据的影响。 |
| 关键词: 励磁涌流 异常数据 相似度识别 Hausdorff距离 |
| DOI：10.7667/PSPC180043 |
| 投稿时间：2018-01-10修订日期：2018-05-10 |
| 基金项目:国家自然科学基金项目资助(51477090);南方电网公司科技项目资助(GZHKJXM20160038) |
|
| A new method of inrush current identification based on partial Hausdorff distance |
| GUO Qianwen,MO Wenxiong,ZHENG Fangqing |
| (Tests and Research Institute of Guangzhou Power Supply Bureau, Guangzhou 510054, China) |
| Abstract: |
| The electromagnetic environment of the substation is more complex, secondary equipment may be affected by electromagnetic interference and other factors, which will produce false data and thus lead to protection misoperation. However, conventional methods of transformer inrush current identification do not take into account the influence of false data interference on protection. By analyzing the influence of false data on different methods, it can be concluded that existing methods may slow-operate even mal-operate. To solve this problem, a new magnetizing inrush identification method based on similarity identification is proposed. The magnetizing inrush current and fault current are identified by comparing the waveform similarity between the detected current and the sinusoidal signal. Simulation results show that this method can identify inrush current quickly and effectively, and it can not be affected by false data to a certain degree. This work is supported by National Natural Science Foundation of China (No. 51477090) and Science and Technology Project of China Southern Power Grid Company (No. GZHKJXM20160038). |
| Key words: inrush current false data similarity identification Hausdorff distance |
