Page 64 - 《应用声学》2025年第2期

第 44 卷 第 2 期 Vol. 44, No. 2
2025 年 3 月 Journal of Applied Acoustics March, 2025

⋄ 研究论文 ⋄

应用于电气设备局部放电定位的改进相位变换

加权可控响应功率算法

李昌伟 1 杨祥国 1† 邓如朝 1 张 辉 2

(1 武汉理工大学船海与能源动力工程学院 武汉 430063)
(2 武汉国能测控有限公司 武汉 430061)
摘要：为了提高相位变换加权可控响应功率 (SRP-PHAT) 算法在低信噪比 (SNR) 情况下对电气设备局部放
电声源的定位性能，提出了一种基于矩阵低秩稀疏分解的改进算法。改进算法将所有传声器对信号的互功率
谱组成广义互谱矩阵，对该互谱矩阵进行低秩稀疏分解得到一个低秩矩阵和一个稀疏矩阵。从低秩矩阵中取
出相应元素作为所有传声器对信号的互功率谱，再由此互功率谱计算可控响应功率 (SRP) 函数值。通过仿真
以及实验对所提算法进行验证，结果表明在低 SNR 情况下，相比于传统 SRP-PHAT 算法，改进算法的方位角
和仰角定位误差分别平均减小了 4.9 和 4.5 。改进后的算法对噪声具有更强的鲁棒性，对电气设备局部放电
◦
◦
定位具有较高的应用价值。
关键词：局部放电；声源定位；传声器阵列；低秩稀疏分解；相位变换加权可控响应功率
中图法分类号: TN912.3 文献标识码: A 文章编号: 1000-310X(2025)02-0324-15
DOI: 10.11684/j.issn.1000-310X.2025.02.007

The improved steered response power-phase transform algorithm applied to
partial discharge localization in electrical equipment

1
1
1
LI Changwei , YANG Xiangguo , DENG Ruzhao and ZHANG Hui 2
(1 School of Naval Architecture, Ocean and Energy Power Engineering, Wuhan University of Technology,
Wuhan 430063, China)
(2 Wuhan GNCK Co., Ltd., Wuhan 430061, China)

Abstract: To enhance the localization performance of the steered response power-phase transform (SRP-
PHAT) algorithm for electrical equipment partial discharge sources in low signal-to-noise ratio (SNR) con-
ditions, an improved algorithm based on matrix low-rank sparse decomposition is proposed. The enhanced
algorithm forms a generalized cross-spectral matrix from the cross-power spectrums of signals from all micro-
phone pairs. This cross-spectral matrix is subjected to a low-rank sparse decomposition, resulting in a low-rank
matrix and a sparse matrix. The relevant elements are extracted from the low-rank matrix to represent the
cross-power spectrums of the signals for all microphone pairs. Subsequently, the steered response power (SRP)
function values are computed from these cross-power spectrums. Simulation and experimental validation of the
proposed algorithm show that, under low SNR conditions, compared to the traditional SRP-PHAT algorithm,
◦
◦
the improved algorithm reduces the errors in azimuth and elevation localization by an average of 4.9 and 4.5 ,
respectively. The enhanced algorithm exhibits stronger robustness against noise, demonstrating its valuable
application in the localization of electrical equipment partial discharge.
2023-10-22 收稿; 2024-01-03 定稿
作者简介: 李昌伟 (1999– ), 男, 湖北随州人, 硕士研究生, 研究方向: 阵列信号处理。
† 通信作者 E-mail: yangxiangguo123@163.com