| 引用本文: | 薛士敏,张君婷,李博通,李翔宇.适用于直流保护的数学形态学降噪与噪声干扰识别方法[J].电力系统保护与控制,2024,52(13):136-148.[点击复制] |
| XUE Shimin,ZHANG Junting,LI Botong,LI Xiangyu.Mathematical morphology denoising and interference recognition method for DC protection[J].Power System Protection and Control,2024,52(13):136-148[点击复制] |
|
| 摘要: |
| 柔性直流输电系统是调节我国能源分布不均问题的重要手段，直流保护是保证柔直系统安全的重要技术。现有工程中多采用行波保护作为主保护，然而行波保护存在易受噪声影响、故障波头不易捕捉等问题，制约了其广泛应用。针对上述问题，提出数学形态学降噪与噪声干扰识别方法，并根据直流保护采样信号特点进行参数设计，用于直流保护采样信号降噪和噪声干扰的识别。首先，确定基于自适应多尺度形态滤波的直流信号降噪方法，并设计结构元素长度和形态学滤波器类型、自适应多尺度形态滤波器的滤波器阶数和步长。其次，确定基于形态学梯度变换的噪声干扰识别方法，并设计形态学梯度变换的阈值。接着，从不同角度对形态学降噪方法与现有典型信号处理方法的滤波效果进行对比分析。最后，对形态学降噪与噪声干扰识别方法和保护的配合效果进行验证。形态学降噪与噪声干扰识别方法所需采样频率较低，运算量适中，滤波效果好，对直流保护适用性好。 |
| 关键词: 行波保护 自适应多尺度形态滤波 噪声干扰识别 形态学梯度 故障波头捕捉 |
| DOI：10.19783/j.cnki.pspc.231197 |
| 投稿时间：2023-09-11修订日期：2023-11-12 |
| 基金项目:国家电网有限公司总部管理科技项目资助(5500- 202199270A-0-0-00) |
|
| Mathematical morphology denoising and interference recognition method for DC protection |
| XUE Shimin,ZHANG Junting,LI Botong,LI Xiangyu |
| (Key Laboratory of Smart Grid of Ministry of Education (Tianjin University), Tianjin 300072, China) |
| Abstract: |
| The flexible DC transmission system is an important means of adjusting the uneven distribution of energy in China, and DC protection is the key technology to ensure the safety of that system. Traveling wave protection is mostly used as the main protection in existing projects. However, the wide use of traveling wave protection on a flexible DC transmission system is limited by serious noise interference and a difficult wave head extraction process. Given these issues, a mathematical morphology denoising and interference recognition method is proposed. The parameters are designed according to the characteristic of the DC protection sampling signal to realize DC protection sampling signal denoising and interference recognition. First, a denoising method based on adaptive multiscale mathematical morphology filter (AM-MMF) is determined, where the scale of structuring elements, the type of morphology filter, the filter order and step size of AM-MMF are designed. Secondly, an interference recognition method based on morphological gradient (MG) transformation is determined, and threshold values of MG are explored. Then, the comparison of the filter effect between mathematical morphology denoising and typical denoising method is made from different perspectives. Finally, the coordination effect of mathematical morphology denoising and interference recognition with protection scheme is validated. Mathematical morphology denoising and interference recognition has superiority in low sampling frequency, moderate computational requirements, excellent filter effect and excellent suitability for DC protection. |
| Key words: traveling wave protection adaptive multiscale mathematical morphology filter interference recognition morphological gradient fault wave head extraction |
