采用多级传感器融合网络的强噪声下高压断路器故障诊断方法研究

王艳新; 张正润; 闫静; 刘志远; 耿英三; 王建华

doi:null

您当前的位置：

首页 >

文章列表页 >

采用多级传感器融合网络的强噪声下高压断路器故障诊断方法研究

更新时间：2026-07-28

- 采用多级传感器融合网络的强噪声下高压断路器故障诊断方法研究
- Fault Diagnosis of High-Voltage Circuit Breakers under Strong Noise Based on a Multi-Level Sensor Fusion Network
- 西安交通大学学报 2026年
- 作者机构：
  
  西安交通大学电工材料电气绝缘全国重点实验室，710049，西安
- 作者简介：
- 基金信息：
- DOI：
  中图分类号： TM561;TP18
- 收稿：2026-06-10，
  
  修回：2026-07-24，
  
  录用：2026-07-28，
- 稿件说明：
移动端阅览
王艳新, 张正润, 闫静, 等. 采用多级传感器融合网络的强噪声下高压断路器故障诊断方法研究[J/OL]. 西安交通大学学报, 2026.

WANG Yanxin, ZHANG Zhengrun, YAN Jing, et al. Fault Diagnosis of High-Voltage Circuit Breakers under Strong Noise Based on a Multi-Level Sensor Fusion Network[J/OL]. JOURNAL OF XI’AN JIAOTONG UNIVERSITY, 2026.
王艳新, 张正润, 闫静, 等. 采用多级传感器融合网络的强噪声下高压断路器故障诊断方法研究[J/OL]. 西安交通大学学报, 2026. DOI：

WANG Yanxin, ZHANG Zhengrun, YAN Jing, et al. Fault Diagnosis of High-Voltage Circuit Breakers under Strong Noise Based on a Multi-Level Sensor Fusion Network[J/OL]. JOURNAL OF XI’AN JIAOTONG UNIVERSITY, 2026. DOI：

摘要

针对强噪声工况下高压断路器机械故障信号易受背景干扰、多源信息利用不足及小样本特征判别能力弱的问题，提出一种基于多级传感器融合网络的诊断方法。该方法从数据级、特征级和决策级三个层面构建多级融合框架，实现多源信息的协同建模与鲁棒诊断。首先，在数据级实现同构传感器信号多通道协同表征，增强原始信号中故障信息的可辨识度。其次，在特征级构建以小波卷积为核心的特征学习网络，对异构传感器信号进行联合时频建模，利用小波卷积在多尺度下抑制背景噪声干扰，保持良好的特征提取稳定性。最后，在决策级基于信息熵对多传感器诊断结果进行可信度加权融合，降低单一传感器误判对整体决策的影响，并通过时间-频率原型对比学习增强类内聚集和类间分离。研究、结果表明：相比最优特征级融合方法，精确率、召回率和 F1 值分别提高了 2.58%、4.85%和 2.70%；在 -15 dB 强噪声下， F1 值仍高于 90%；相比传统对比学习模型，3项指标分别提高了 6.52%、9.24%和 6.87%，从而验证了方法的鲁棒性与小样本诊断的有效性。该研究为实现强噪声与小样本条件下高压断路器机械故障的诊断提供了一种有效解决方法。

Abstract

To address the issues of mechanical fault signals of high-voltage circuit breakers being susceptible to background interference under strong noise conditions

insufficient utilization of multi-source information

and weak discriminative capability under small-sample conditions

a diagnostic method based on a multi-level sensor fusion network is proposed. This method constructs a multi-level fusion framework at three levels—data level

feature level

and decision level—to achieve collaborative modeling and robust diagnosis of multi-source information. First

at the data level

multi-channel collaborative representation of homogeneous sensor signals is realized to enhance the identifiability of fault information in the raw signals. Second

at the feature level

a feature learning network centered on wavelet convolution is built to perform joint time–frequency modeling of heterogeneous sensor signals. The wavelet convolution suppresses background noise interference at multiple scales while maintaining stable feature extraction. Finally

at the decision level

credibility-weighted fusion of multi-sensor diagnostic results is performed based on information entropy

reducing the impact of misjudgment by a single sensor on the overall decision. In addition

time–frequency prototype contrastive learning is employed to enhance intra-class compactness and inter-class separation. The results shows that

compared with the optimal feature-level fusion method

the precision

recall

and F1-score are improved by 2.58%

4.85%

and 2.70%

respectively. Under strong noise at −15 dB

the F1-score remains above 90%. Compared with traditional contrastive learning models

the three metrics are increased by 6.52%

9.24%

and 6.87%

respectively

verifying the robustness and small-sample diagnostic effectiveness of the proposed method. This study provides an effective solution for fault diagnosis of high-voltage circuit breakers under strong-noise and small-sample conditions.

关键词

Keywords

references

Zhang Z , Wang Y , Yan J , et al . Multi‐Information Fusion Diagnosis of High‐Voltage Circuit Breakers via Newton–Raphson Optimised Transformer and Meta‐Transfer Learning [J ] . IET Generation , Transmission & Distribution, 2026 , 20 ( 1 ): e70252 .

赵陈琛 , 张国钢 , 刘洁 , 等 . 基于物理信息和迁移网络的高压断路器弹簧操动机构机械参数辨识及故障诊断 [J/OL ] . 电工技术学报 , 1 - 13 [ 2025-08-01 ] .

Zhao Chenchen , Zhang Guogang , Liu Jie , et al . Mechanical parameter identification and fault diagnosis of spring operating mechanism of high-voltage circuit breaker based on physical information and transfer network [J/OL ] . Transactions of China Electrotechnical Society , 2025 .

范兴明 , 许洪华 , 张思舜 , 等 . 基于改进蜣螂优化算法深度混合核极限学习机的高压断路器故障诊断 [J ] . 电工技术学报 , 2025 , 40 ( 12 ): 3994 - 4003 .

Fan Xing ming , Xu Honghua , Zhang Sishun , et al . Fault diagnosis of high voltage circuit breaker based on improved dung beetle optimizer algorithm deep hybrid kernel extreme learning machine [J ] . Transactions of China Electrotechnical Society , 2025 , 40 ( 12 ): 3994 - 4003 .

范兴明 , 许洪华 , 张思舜 , 等 . 基于IDBO-DHKELM的高压断路器故障诊断研究 [J/OL ] . 电工技术学报 , 1 - 12 [ 2025-08-01 ] .

Fan Xing ming , Xu Honghua , Zhang Sishun , et al . Research on fault diagnosis of high voltage circuit breaker based on IDBO-DHKELM [J/OL ] . Transactions of China Electrotechnical Society , 2025 .

王艳新 , 闫静 , 荆乾震 , 等 . 基于先验知识约束和对比域适应学习的高压断路器机械故障诊断方法研究 [J ] . 电工技术学报 , 1 - 13 [ 2025-10-30 ] .

Wang Yanxin , Yan Jing , Jing Qianzhen , et al . Mechanical fault diagnosis of high-voltage circuit breakers based on prior knowledge constraints and contrastive domain adaptation [J/OL ] . Transactions of China Electrotechnical Society , 2025 .

Wang Y , Yan J , Zhang Z , et al . Few-Shot Mechanical Fault Diagnosis of High-Voltage Circuit Breakers Using Prior Knowledge-Constrained Wavelet-Based Multi-Frequency Feature Fusion [C ] // 2025 IEEE China International Youth Conference on Electrical Engineering (CIYCEE) . IEEE , 2025 : 1 - 7 .

Song Q , Wang J , Song Q , et al . Fault Diagnosis for the Operating Mechanism of High-Voltage Circuit Breakers using Multi-Channel Fusion Convolutional Neural Network [J ] . 2024 .

Ye X , Yan J , Wang Y , et al . A novel circuit breaker fault diagnosis method based on dense residual and attention mechanism [J ] . IET Generation , Transmission & Distribution, 2023 , 17 ( 19 ): 4316 - 4328 .

Wang Yanxin , Yan Jing , Wang Jianhua , et al . A novel hybrid transfer learning approach for small-sample high-voltage circuit breaker fault diagnosis on-site [J ] . IEEE Transactions on Industry Applications , 2023 , 59 ( 4 ): 4942 - 4950 .

Wang Yanxin , Yan Jing , Ye Xinyu , et al . Few-shot transfer learning with attention mechanism for high-voltage circuit breaker fault diagnosis [J ] . IEEE Transactions on Industry Applications , 2022 , 58 ( 3 ): 3353 - 3360 .

Shao Y , Zhang Z , Wu J , et al . Multi-Sensor Sensitivity Assessment Strategy for High-Voltage Circuit Breaker Fault Diagnosis Using GA-SoftMax Model [J ] . IEEE Transactions on Power Delivery , 2025 .

张国宝 , 王朝廷 , 黄伟民 , 等 . 基于多源传感器数据融合的断路器故障诊断方法 [J ] . 高电压技术 , 2025 , 51 ( 02 ): 660 - 668 .

Zhang Guobao , Wang Zhaoyan , Huang Weimin , et al . Fault Diagnosis Method of Circuit Breaker Based on Multi-source Sensor Data Fusion [J ] . High Voltage Engineering , 2025 , 51 ( 02 ): 660 - 668 .

Wang Y , Yan J , Qi M , et al . A Novel Meta-Learning and Network Architecture Search Approach for Few-Shot High-Voltage Circuit Breaker Fault Diagnosis [C ] // 2023 IEEE 6th International Electrical and Energy Conference (CIEEC) . IEEE , 2023 : 122 - 127 .

张瑞霖 , 张悦 , 孙晓兰 , 等 . 基于可变形卷积与自监督对比学习的GIS局部放电诊断方法 [J ] . 高电压技术 , 2024 , 50 ( 11 ): 5022 - 5033 .

Zhang Ruilin , Zhang Yue , Sun Xiaolan , et al . GIS partial discharge diagnosis method based on deformable convolution and self-supervised contrastive learning [J ] . High Voltage Engineering , 2024 , 50 ( 11 ): 5022 - 5033 .

杨夏洁 , 林子谦 , 徐丹丹 , 等 . 基于联邦对比学习的航天器故障诊断研究 [J/OL ] . 计算机应用研究 , 1 - 8 [ 2025-08-01 ] .

Yang Xiajie , Lin Ziqian , Xu Dandan , et al . Study of spacecraft fault diagnosis based on federated comparative learning [J/OL ] . Application Research of Computers , 2025 .

张珂 , 郑朝烨 , 石超君 , 等 . 基于多模态对比学习的输电线路螺栓缺陷分类 [J ] . 高电压技术 , 2025 , 51 ( 02 ): 630 - 641 .

Zhang Ke , Zhe Chaoye , Shi Chaojun , et al ． Transmission line bolt defects classification based on multi-modal contrastive learning [J ] . High Voltage Engineering , 2025 , 51 ( 02 ): 630 - 641 .

Yan J , Wang Y , Yang Z , et al . Few-shot mechanical fault diagnosis for a high-voltage circuit breaker via a transformer–convolutional neural network and metric meta-learning [J ] . IEEE Transactions on instrumentation and measurement , 2023 , 72 : 1 - 11 .

Yang Chaoying , Liu Jie , Xu Qi , et al . A generalized graph contrastive learning framework for few-shot machine fault diagnosis [J ] . IEEE Transactions on Industrial Informatics , 2023 , 20 ( 2 ): 2692 - 2701 .

Miao Y , Wang J , Zhang B , et al . Practical framework of Gini index in the application of machinery fault feature extraction [J ] . Mechanical Systems and Signal Processing , 2022 , 165 : 108333 .

Chen Z , Yang Y , He C , et al . Feature extraction based on hierarchical improved envelope spectrum entropy for rolling bearing fault diagnosis [J ] . IEEE Transactions on Instrumentation and Measurement , 2023 , 72 : 1 - 12 .

Fan C , Yun M , Chen L , et al . DWMFusion: A lightweight medical image fusion network based on discrete wavelet convolution and selective state space models [J ] . Expert Systems with Applications , 2025 , 286 : 128040 .

Xiao X , Li C , He H , et al . Rotating machinery fault diagnosis method based on multi-level fusion framework of multi-sensor information [J ] . Information Fusion , 2025 , 113 : 102621 .

Wang Y , Yan J , Yang Z , et al . A domain adaptive deep transfer learning method for gas-insulated switchgear partial discharge diagnosis [J ] . IEEE Transactions on Power Delivery , 2021 , 37 ( 4 ): 2514 - 2523 .

Dong Y , Jiang H , Wang X , et al . An interpretable integration fusion time-frequency prototype contrastive learning for machine fault diagnosis with limited labeled samples [J ] . Information Fusion , 2025 , 124 : 103340 .

Bettahar T , Rahmoune C , Benazzouz D , et al . New method for gear fault diagnosis using empirical wavelet transform, Hilbert transform, and cosine similarity metric [J ] . Advances in Mechanical Engineering , 2020 , 12 ( 6 ): 1687814020927208 .

Wang Y , Yan J , Zhang Z , et al . A novel multi-sensor fusion method for diagnosing insulation defects in gas-insulated substations guided by adaptive-attention and contrastive-based few-shot learning [J ] . Applied Soft Computing , 2026 : 114600 .

Rathore A , Gupta P , Sharma R , et al . Day ahead solar forecast using long short term memory network augmented with Fast Fourier transform-assisted decomposition technique [J ] . Renewable Energy , 2025 , 247 : 123021 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

电力变压器故障诊断多智能体大模型构建与应用

采用双小波字典的齿轮箱复合故障特征提取方法

一维卷积神经网络在机械故障特征提取中的可解释性研究

蜉蝣优化双通道网络在齿轮箱故障诊断中的应用

一种双阻尼小波赋能的可解释卷积神经网络在轴承故障诊断中的应用