中圖分類(lèi)號(hào):TM855 文獻(xiàn)標(biāo)志碼:A DOI: 10.16157/j.issn.0258-7998.234499 中文引用格式: 羅日平,羅穎婷,賴詩(shī)鈺,等. 基于EEMD奇異值熵的局部放電模式識(shí)別[J]. 電子技術(shù)應(yīng)用,2024,50(3):53-58. 英文引用格式: Luo Riping,Luo Yingting,Lai Shiyu,et al. Partial discharge pattern recognition based on EEMD singular value entropy[J]. Application of Electronic Technique,2024,50(3):53-58.
Partial discharge pattern recognition based on EEMD singular value entropy
Luo Riping1,Luo Yingting2,Lai Shiyu2,Zhao Xianyang3,Wang Liqi4
1.China Southern Power Grid Scientific Research Institute Co., Ltd., Guangzhou 510700, China; 2.Electirc Power Research Insitute of Guangdong Power Grid Co., Ltd., Guangzhou 510080, China; 3.State Grid Shandong Electric Power Company Heze Power Supply Co., Ltd., Heze 274000, China; 4.School of Electronics and Information Engineering, Shanghai University of Electric Power, Shanghai 201306, China
Abstract: Aiming at the non-stationary of gas insulatede switchgear(GIS) partial discharge fault signal and the low accuracy of discharge type recognition, a partial discharge pattern recognition method based on ensemble empirical mode decomposition(EEMD) singular value entropy is proposed. Firstly, the EEMD algorithm is used to decompose the original signals of partial discharge to intrinsic mode functions(IMFs), according to the mean square error, kurtosis and euclidean distance evaluation index, the optimal modal component with most implicit discharge information is selected for signal reconstruction. Secondly, the singular value decomposition is performed on the reconstructed signal, and the singular value entropy is calculated in combination with the information entropy algorithm. Finally, according to the singular value entropy, the type of GIS partial discharge is distinguished. The experiment results show that by comparing with the traditional EMD singular value entropy and VMD singular value entropy algorithms, the method in this paper can effectively identify the discharge type through the singular entropy values in different intervals.
Key words : EEMD;singular value entropy;evaluation index;partial discharge;pattern recognition