作者：邵宇鹰¹, 王枭², 彭鹏¹, 袁国刚², 柯楠¹

作者单位：1. 国网上海市电力公司，上海 200122;
2. 上海睿深电子科技有限公司，上海 201108

关键词：声成像技术;电力设备缺陷;异响检测;波束形成算法;DAMAS算法

摘要：

近年来变电站内在运电气设备异响现象时有发生，其中某些异响情况极易导致设备故障，有必要及时发现并处理，目前在变电站实际工作中，运维人员多采用耳听的方式判断异响位置。针对上述问题，该文分析变电站内电气设备异响来源，介绍基于DAMAS算法的声成像技术，研制一种可搭载于变电站巡检机器人的电力设备异响检测系统，研究GIS设备机械振动异响、高压端电晕放电异响两种变电站典型设备异响情况。研究结果表明：该方法可快速准确地判断两种异响位置，基于DAMAS算法的声成像技术可有效提升传统波束形成算法的定位精度，定位距离可达30 m以上，再结合异响信号图谱特征和异响位置结构信息，可初步判断缺陷类型，满足变电站内异响缺陷定位需求。

Research on defect detection method of power equipment based on acoustic imaging technology
SHAO Yuying¹, WANG Xiao², PENG Peng¹, YUAN Guogang², KE Nan¹
1. State Grid Shanghai Municipal Electric Power Company, Shanghai 200122, China;
2. Shanghai Rhythm Electronic Technology Co., Ltd., Shanghai 201108, China
Abstract: Abnormal noise of electrical equipment in substation occurs now and then. Some of the abnormal sound conditions can easily lead to equipment failure. Therefore, It is necessary to find and deal with it in time. At present, substation operation and maintenance personnel mostly use ear to judge abnormal sound position. Aiming at these problems above, this paper analyzes the source of abnormal noise of electrical equipment in substation. Acoustic imaging technology based on DAMAS algorithm is introduced. An automatic abnormal sound detection system for power equipment which can be equipped with substation patrol robot is developed. The abnormal response of GIS equipment and the abnormal response of high-voltage corona discharge are studied. The results suggested that the method can quickly and accurately determine the location of abnormal sound. Acoustic imaging technology based on DAMAS algorithm can effectively improve the positioning accuracy of traditional beamforming algorithms. The positioning range is over 30 meters. Combined with the map characteristics and location structure information of abnormal response signal, the fault type can be preliminarily judged. The requirements of substation internal abnormal sound fault location can satisfy.
Keywords: acoustic imaging technology;power equipment failure;abnormal sound detection;beamforming;DAMAS algorithm
2021, 47(7):42-48  收稿日期: 2020-07-13;收到修改稿日期: 2020-09-01
基金项目: 国网上海市电力公司科技项目资助（52097018000F）
作者简介: 邵宇鹰(1977-),男,江苏常州市人,高级工程师,博士,主要从事电力设备状态监测和新能源技术研究与应用
参考文献
[1] 牛博, 唐鑫, 张欣宜, 等. 气体绝缘组合电器中电压互感器振动特性研究[J]. 高压电器, 2019, 55(11): 74-78
[2] 蒋玲, 曲全磊, 王志惠, 等. 高温差工况下长母线GIS设备振动特性研究[J]. 高压电器, 2019, 55(11): 144-151
[3] WERLE P, BORSI H, GOCKENBACH E. Diagnosing the insulation condition of dry-type transformers using a multiple sensor partial discharge localization technique[C]//Conference Record of the 2002 IEEE International Symposium on Electrical Insulation, 2002.
[4] 高燃, 尹建军, 刘文海, 等. 现代数字滤波技术在大型电力变压器异响诊断中的应用[J]. 变压器, 2015, 52(12): 77-80
[5] 曹浩, 吴晓文, 卢玲, 等. 基于波束形成声成像技术的某±800 kV换流站噪声源识别[J]. 高压电器, 2019, 55(11): 171-176
[6] 余长厅, 黎大健, 陈梁远, 等. 基于声纹及振动的变压器故障诊断技术研究[J]. 高压电器, 2019, 55(11): 248-254
[7] 李炼炼, 孟刚, 邓慰, 等. 输变电设备电晕放电紫外图谱量化参数提取[J]. 高压电器, 2017, 53(12): 229-235
[8] 刘媛, 贾勇勇, 杨景刚, 等. 一起220 kV GIS设备异响缺陷分析与处理[J]. 高压电器, 2020, 56(2): 246-251
[9] 王昕, 赵飞, 蒋佐富, 等. 迁移学习和卷积神经网络电力设备图像识别方法[J]. 中国测试, 2020, 46(5): 108-113
[10] GOCKENBACH E, WERLE P, BORSI H. Monitoring and diagnostic system for dry-type transformers[C]//IEEE 7th International Conference on Solid Dielectrics, 2001.
[11] JACOB J C, JORGEN H Improvement of cross spectral beamforming[C]// Proceedings of Internoise, 2003.
[12] CHRISTANSEN J J, HALD J. Beamforming[R]. B&K Technical Review, 2004(1): 1-31.
[13] HUANG X, BAI L, VINOGRADOV I, et al. Adaptive beamforming for array signal processing in aero acoustic measurements[J]. The Journal of the Acoustical Society of America, 2012, 131(3): 2152-2161
[14] BROOKS T F, HUMPHREYS W M. A deconvolution approach for the mapping of acoustic sources (DAMAS) determined from phased microphone arrays[J]. Journal of Sound and Vibration, 2006, 294(4): 856-879