[1]江 楠,杨 婷,周怡娜,等.融合电流-电磁应力特征的 XGBoost-SHAP 潜油电泵故障诊断[J].机械与电子,2026,44(05):56-63.
　JIANG Nan,YANG Ting,ZHOU Yina,et al.Fault Diagnosis of XGBoost-SHAP Submersible Electric Pump Based on Current-electromagnetic Stress Feature Fusion[J].Machinery & Electronics,2026,44(05):56-63.
点击复制

融合电流-电磁应力特征的 XGBoost-SHAP 潜油电泵故障诊断()
分享到：

《机械与电子》[ISSN:1001-2257/CN:52-1052/TH]

卷:
44
期数:
2026年05期
页码:
56-63
栏目:
智能检测
出版日期:
2026-05-27

文章信息/Info

Title:
Fault Diagnosis of XGBoost-SHAP Submersible Electric Pump Based on Current-electromagnetic Stress Feature Fusion
文章编号:
1001-2257 ( 2026 ) 05-0056-08
作者:
江 楠 1 ; 杨 婷 1 ; 周怡娜 1 ; 支继强 2 ; 冷 池 1
1. 东北石油大学电气信息工程学院,黑龙江 大庆 163318 ;
2. 东北石油大学石油工程学院,黑龙江 大庆 163318
Author(s):
JIANG Nan1 ;  YANG Ting1 ;  ZHOU Yina1 ;  ZHI Jiqiang2 ;  LENG Chi1
( 1.School of Electrical and Information Engineering , Northeast Petroleum University , Daqing 163318 , China ;
2.School of Petroleum Engineering , Northeast Petroleum University , Daqing 163318 , China )
关键词:
潜油电泵; 故障诊断; XGBoost ;  SHAP
Keywords:
electric submersible pump ;  fault diagnosis ;  XGBoost ;  SHAP
分类号:
TP181 ;TE933
文献标志码:
A
摘要:
针对海上油田潜油电泵工况复杂、故障类型多且传统诊断可解释性差的问题,以某海上油田潜油电泵为对象,采集正常工况及 5 种故障工况下的定子电流、电磁应力信号,采用滑动窗口与标准化处理,构建包含时域、频域和电 磁交互量的 47 维融合特征。基于该特征集建立 XGBoost 等多种分类模型并比较性能。结果表明,树集成模型表现最好,其中 XGBoost 和随机森林测试准确率分别为 97.9% 和 95.7% ,明显优于 BP 神经网络。进一步结合 SHAP 对 XGBoost 进行全局与局部解释,识别出电流频谱标准差、波形因子及电磁应力频谱能量等关键特征,揭示不同工况下电 磁响应差异。研究结果表明,基于 XGBoost-SHAP 的融合特征诊断方法兼具较高精度和可解释性,可为潜油电泵智能运维提供理论与技术支持。
Abstract:
Aiming at the problems of complex working conditions , fault types and poor interpretability of traditional diagnosis of submersible electric pumps in offshore oilfields , this paper takes an offshore oilfield submersible electric pump as the object , collects stator current and electromagnetic stress signals under normal working conditions and five fault conditions , and uses sliding window and standardization processing to construct 47-dimensional fusion features including time domain , frequency domain and electro magnetic interaction.Based on this feature set , classification models such as XGBoost are established and their performance is compared.The results show that the tree ensemble model , and the accuracy of XGBoost and random forest test is about 97.9% and 95.7% , respectively , which is significantly better than that of BP neural network.he global and local interpretation of XGBoost is further combined with SHAP to identify key features such as current spectrum standard deviation , waveform factor and electromagnetic stress spectrum energy , reveal the difference of electro magnetic response under different working conditions.The research shows that the fusion feature diagnosis method based on XGBoost-SHAP has both high accuracy and interpretability , provid theoretical and technical support for the intelligent operation and maintenance of submersible electric pump.

参考文献/References:

[ 1 ] 高生 . 潜油电泵机组失效原因分析与改进[ J ] . 中国石油和化工标准与质量,2025 , 45 ( 2 ): 38-41.

[ 2 ] 陈亮 . 海上平台采油井电潜泵机械故障信号提纯研究[ J ] . 机械与电子, 2023 , 41 ( 6 ): 71-75.
[ 3 ] Li Qiang , Li Kang , Gao Xiaoyong , et al.Anomaly detection based on temporal attention network with adaptive threshold adjustment for electrical submersible pump [ J ] .IEEE Transactions on Instrumentation and Measurement , 2024 , 73 : 3526214.
[ 4 ] 杜银昌,袁杰 . 基于完整信息主成分分析的电潜泵故障检测研究[ J ] . 自动化应用, 2025 , 66 ( 17 ): 124-127.
[ 5 ] 周逸飞,刘新福,曹砚锋,等 . 基于 LSTM-PNN 神经网络的电潜泵故障诊断方[ J ] . 机床与液压, 2024 , 52( 19 ): 209-215.
[ 6 ] 杨岗,卫昱乾,邓琴,等 . 机器学习驱动的牵引电机轴承故障特征增强算法研究[ J ] . 中国测试, 2025 , 51 ( 12 ):16-24.
[ 7 ] 姜凯华,张永,谢迎谱,等 . 基于深度森林和 SMOTE 算法的输电线路故障识别方法[ J ] . 浙江电力,2025 , 44( 12 ): 125-136.
[ 8 ] Alhams A , Abdelhadi A , Badri Y , et al.Enhanced bearing fault diagnosis through trees ensemble method and feature importance analysis [ J ] .Journal of Vibration Engineering and Technologies , 2024 , 12 : 109-125.
[ 9 ] 张雅晖,李天乐 . 锦州 25-1 海上平台电潜泵故障模式识别与智能检测算法数据集[ DB / OL ] .V1. 国家基础学科公共科学数据中心( 2024-06-24 )[ 2026-01-02 ] .https : ∥cstr.cn / 16666.11.nbsdc.awb1r9mv.

相似文献/References:

[1]吕明珠１,２,苏晓明 １,等. 改进粒子群算法优化的支持向量机在滚动轴承故障诊断中的应用[J].机械与电子,2019,(01):42.
　,,et al.Application of SVM Optimized by IPSO in Rolling Bearing Fault Diagnosis[J].Machinery & Electronics,2019,(05):42.
[2]汪 伟,李 军,翟旭升,等.某型航空发动机中介主轴承早期微弱故障诊断研究[J].机械与电子,2019,(06):6.
　,,et al.Research on the Early Weak Fault Diagnosis of Aero-engine Intershaft Bearing[J].Machinery & Electronics,2019,(05):6.
[3]赵蕾,傅攀,胡龙飞,等.FOA-WPT降噪和PSO-SVM在滚动轴承故障诊断中的应用[J].机械与电子,2018,(12):3.
　ZHAO Lei,FU Pan,HU Longfei,et al.Applications of FOA-WPT and PSO-SVM in Faults Diagnosis of Rolling Bearing[J].Machinery & Electronics,2018,(05):3.
[4]吕明珠 苏晓明 陈长征 刘世勋.小波包能量熵与EMD结合分析法在风机滚动轴承故障诊断中的应用[J].机械与电子,2018,(06):8.
　LYU Mingzhu,SU Xingming,CHEN Changzheng,et al.Application of Wavelet Packet Energy Entropy and EMD Conjoint Analysis in Fault Diagnosis of Wind Turbine Bearing[J].Machinery & Electronics,2018,(05):8.
[5]唐志宇,毛 兴,王路军.电力变压器绕组变形检测技术研究[J].机械与电子,2018,(11):53.
　TANG Zhiyu,MAO Xing,WANG Lujun.Research on Power Transformer Winding Deformation Detection Technology[J].Machinery & Electronics,2018,(05):53.
[6]杨伟力,于阳阳,罗达灿.基于小波包和PSO-Elman神经网络的滚动轴承故障诊断[J].机械与电子,2016,(05):13.
　YANG Weili,YU Yangyang,LUO Dacan.Rolling Bearing Fault Diagnosis Using Wavelet Packet Analysis and PSO-Elman Neural Network[J].Machinery & Electronics,2016,(05):13.
[7]袁小凯,李 果,蒋屹新,等.基于 ＨＳ- ＥＬＭ 的油浸式变压器故障诊断[J].机械与电子,2019,(12):30.
　,,et al.Fault Diagnosis of Oil Immersion Transformer Based on HS-ELM Algorithm[J].Machinery & Electronics,2019,(05):30.
[8]孟庆宇.基于 ＥＥＭＤ- ＳＶＭ 的自动扶梯主驱动轴轴承故障诊断方法研究[J].机械与电子,2020,(05):51.
　Research on Fault Diagnosis Method for Main Drive Shaft Bearing of Escalator Based on EEMD-SVM [J].Machinery & Electronics,2020,(05):51.
[9]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,(05):67.
[10]齐爱玲１,李 琳１,朱亦轩２,等.基于融合特征的双通道CNN滚动轴承故障识别[J].机械与电子,2021,(05):15.
　QI Ailing,LI Lin,ZHU Yixuan,et al.Dual Channel CNN Bearing Fault Identification Based on Fusion Feature[J].Machinery & Electronics,2021,(05):15.

备注/Memo

备注/Memo:
收稿日期: 2026-01-09
基金项目:国家自然科学基金青年科学基金项目( C 类, 62503106 )
作者简介:江 楠 ( 1982- ),女,黑龙江大庆人,副教授,研究方向为新能源系统优化控制、故障诊断及容错控制;杨 婷 ( 2000- ),女,湖北恩施人,硕士研究生,研究方向为电工理论与新技术;周怡娜 ( 1989- ),女,黑龙江大庆人,讲师,研究方向为管道故障诊断相关技术、机器学习和自然语言处理;支继强 ( 1985- ),男,黑龙江大庆人,博士,副教授,研究方向为采油采气工程理论与技术;冷 池 ( 2004- ),女,河南新乡人,硕士研究生,研究方向为电工理论与新技术。
更新日期/Last Update: 2026-08-25