[1]郑 宏,鲍美军,卢衍泓,等.基于多测点振动特征的变压器负载运行状态评估方法[J].机械与电子,2026,44(07):93-97.
　ZHENG Hong,BAO Meijun,LU Yanhong,et al.Evaluation Method of Transformer Load Operation Status Based on Multi-measurement Point Vibration Characteristics[J].Machinery & Electronics,2026,44(07):93-97.
点击复制

基于多测点振动特征的变压器负载运行状态评估方法()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年07期
页码:
93-97
栏目:
电力控制
出版日期:
2026-07-25

文章信息/Info

Title:
Evaluation Method of Transformer Load Operation Status Based on Multi-measurement Point Vibration Characteristics
文章编号:
1001-2257(2026)07-0093-05
作者:
郑 宏1; 鲍美军1; 卢衍泓1; 孙文星2; 李 丽2; 郭胡森3; 万书亭3
1.杭州柯林电气股份有限公司,浙江 杭州 310015;
2.广东电网有限责任公司,广东 广州 510080;
3.华北电力大学河北省电力机械装备健康维护与失效预防重点实验室,河北 保定 071003
Author(s):
ZHENG Hong1; BAO Meijun1; LU Yanhong1; SUN Wenxing2; LI Li2; GUO Husen3; WAN Shuting3
(1.Hangzhou Kelin Electric Co.,Ltd.,Hangzhou 310015,China;
2.Guangdong Power Grid Co.,Ltd.,Guangzhou 510080,China;
3.Hebei Key Laboratory of Electric Machinery Health Maintenance and Failure Prevention,North China ElectricPower University,Baoding 071003,China)
关键词:
变压器; 振动信号; 功率谱熵; 主成分分析; 多测点
Keywords:
transformer; vibration signal; power spectrum entropy; principal component analysis; multi-measurement point
分类号:
TM411
文献标志码:
A
摘要:
针对变压器负载通常由电流或功率来衡量,严重依赖于电气特性,提出了一种集成多测点振动特征的变压器负载运行状态评估方法。首先,采用改进小波阈值去噪算法对变压器多测点振动信号进行预处理,有效剔除了与变压器负载无关的噪声。然后提取不同时间段振动信号的功率谱熵值,采用主成分分析法将多测点熵值的最大主成分作为最终特征。与实际负载变化进行对比,实验结果表明所提方法误差小于1.98%,同时与单一测点和不同提取算法特征对比,所提方法能够有效表征变压器负载运行状态。
Abstract:
Conventional transformer load assessment relies primarily on electrical quantities such as current or power,and is thus heavily dependent on electrical characteristics.This paper proposes a method that integrates multi-measurement point vibration features for evaluating transformer load operationstate.Firstly,an improved wavelet threshold denoising algorithm is adopted to preprocess the vibration signals acquired from multi measurement points on the transformer,effectively removing the noise irrelevant to the transformer load.Subsequently,the power spectrum entropy of the vibration signals over different time intervals is extracted,and the principal component analysis is applied to take the first principal component of the multi measurement point entropy values as the final feature.Compared with the actual load variations,the experimental results show that the error of the proposed method is less than 1.98%.Meanwhile, compared with features obtained from a single measurement point and those derived using alternative extraction algorithms,the proposed method can effectively characterize the transformer load operation status.

参考文献/References:

[1] 冯洋,尹松,闫敬东,等.基于数字孪生的电网变压器故障诊断方法[J].机械与电子,2022,40(6):26-30.

[2] 王峰.基于模糊神经网络的电力变压器绝缘状态预测方法[J].机械与电子,2022,40(3):50-53.
[3] 张仕民,丰遥.基于振动信号的变压器铁心与绕组故障区分方法[J].电测与仪表,2021,58(6):161-166.
[4] 翟春雨,吴书煜,李颖斌,等.基于小波能量占比与时序预测的换流变压器机械状态智能诊断技术[J].高压电器,2024,60(11):68-76.
[5] 邓旭东,张占龙,夏远灿,等.电力变压器绕组变形状态下的振动信号特性分析[J].电工电能新技术,2024,43(10):13-23.
[6] 宋金伟,宣东海,王维佳,等.基于声纹特征的储能型变压器运维检测技术[J].储能科学与技术,2024,13(8):2758-2760.
[7] Nezhad E A ,Samimi H M .A review of vibration based techniques for the condition assessment and failure detection of transformers[J].Journal of Vibration Engineering and Technologies,2025,13:95.
[8] 吴书煜,汲胜昌,孙建涛,等.在运换流变压器振动监测及其变化规律[J].高电压技术,2022,48(4):1561-1570.
[9] 江翼,程林,陈清淼,等.考虑负载变化的植物绝缘油变压器温升特性研究[J].高压电器,2018,54(6):132-137.
[10] 杨超.变压器不同负载下的电磁损耗分析[J].自动化应用,2024,65(2):111-112.
[11] 于文旭,关向雨,赵俊义,等.基于格子玻耳兹曼的油浸式变压器瞬态温升模拟与负载能力评估[J].电工技术学报,2025,40(10):3315-3325.
[12] 范强,文屹,吕黔苏,等.基于主变负载状况的变压器负荷预测和风险评估[J].电测与仪表,2024,61(11):99-106.
[13] 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.
[14] 栾某德,甘德强,李振垚,等.基于模型响应功率谱密度预测的强迫振荡扰动源设备定位法[J].电力系统自动化,2022,46(22):111-119.
[15] 顾兆军,叶经纬,刘春波,等.基于核主成分分析的半监督日志异常检测模型[J].江苏大学学报(自然科学版),2025,46(1):64-72.

相似文献/References:

[1]王 磊,胡文娟.基于传感器技术的振动智监测研究[J].机械与电子,2020,(10):61.
　WANG Lei,HU Wenjuan.Research on Vibration Intelligent Monitoring Based on Sensor Technology[J].Machinery & Electronics,2020,(07):61.
[2]庄爱军.基于WA-ESN的建筑起重机械故障检测[J].机械与电子,2021,(01):67.
　ZHUANG Aijun.Fault Detection of Construction Crane Based on WA-ESN[J].Machinery & Electronics,2021,(07):67.
[3]王 博,钟 硕,郑晓亮.智能变电站隔离开关机械状态自动监测方法研究[J].机械与电子,2023,41(07):47.
　WANG Bo,ZHONG Shuo,ZHENG Xiaoliang.Research on Automatic Monitoring Method of Mechanical State of Disconnector Switch in Intelligent Substation[J].Machinery & Electronics,2023,41(07):47.
[4]谭继勇,罗 俊,谢江涛,等.基于鲸鱼优化和批量规范化卷积神经网络的振动信号去噪[J].机械与电子,2024,42(04):3.
　TAN Jiyong,LUO Jun,XIE Jiangtao,et al.A Convolutional Neural Network with Whale Optimization and Batch Normalization for the Denoising of Vibration Signal[J].Machinery & Electronics,2024,42(07):3.
[5]许洪华,陈旖旎,丁晓森,等.基于机电融合信号与 1DCNN-MHAM-SVM 的 GIS 局部放电故障诊断[J].机械与电子,2025,(06):37.
　XU Honghua,CHEN Yini,DING Xiaosen,et al.GIS Partial Discharge Fault Diagnosis Based on Electromechanical Signal Fusion and 1DCNN-MHAM-SVM[J].Machinery & Electronics,2025,(07):37.
[6]邹光涛,龚石磊.基于油中溶解气体含量特征的变压器绝缘故障XGBoost识别算法[J].机械与电子,2026,44(03):96.
　ZOU Guangtao,GONG Shilei.XGBoost Identification Algorithm for Transformer Insulation Faults Based on the Characteristics of Dissolved Gas Content in Oil[J].Machinery & Electronics,2026,44(07):96.

备注/Memo

备注/Memo:
收稿日期:2025-10-10
基金项目:国家自然科学基金资助项目(52275109)
作者简介:郑 宏 (1979-),男,浙江衢州人,高级工程师,研究方向为电气设备状态监测与故障诊断;万书亭 (1970-),男,山西长治人,博士,教授,博士研究生导师,研究方向为电力设备状态监测与故障诊断,通信作者,E-mail:52450809@ncepu.edu.cn。
更新日期/Last Update: 2026-08-27