基于改进型级联宽度学习的采煤机截割部齿轮箱故障诊断

李鑫; 李淑华; 陈浩; 司垒; 魏东; 邹筱瑜

doi:10.13272/j.issn.1671-251x.2024110057

基于改进型级联宽度学习的采煤机截割部齿轮箱故障诊断

李鑫^1,,
李淑华¹,
陈浩¹,
司垒¹,
魏东¹,
邹筱瑜^{1, 2}

1.
中国矿业大学机电工程学院，江苏徐州　221116
2.
智能采矿装备技术全国重点实验室，江苏徐州　221116

基金项目: 国家自然科学基金项目（52404178）；江苏省自然科学基金项目（BK20231064）。

详细信息

作者简介:
李鑫（1993—），男，山东滨州人，讲师，博士，主要从事机电装备智能运维研究工作，E-mail：li_xin@cumt.edu.cn

中图分类号: TD67
计量
- 文章访问数: 19
- HTML全文浏览量: 3
- PDF下载量: 1
出版历程
- 收稿日期: 2024-11-18
- 修回日期: 2025-02-24
- 网络出版日期: 2025-02-17
- 刊出日期: 2025-03-14

Fault diagnosis of shearer cutting unit gearbox based on improved cascaded broad learning

LI Xin^1,,
LI Shuhua¹,
CHEN Hao¹,
SI Lei¹,
WEI Dong¹,
ZOU Xiaoyu^{1, 2}

1.
School of Mechanical and Electrical Engineering, China University of Mining and Technology, Xuzhou 221116, China
2.
National Key Laboratory of Intelligent Mining Equipment Technology, Xuzhou 221116, China

摘要

摘要:
采煤机截割部齿轮箱振动监测数据结构复杂，且易出现类别不平衡问题，导致现有基于传统机器学习的智能故障诊断方法易出现错报现象，而基于深度学习的诊断方法模型结构复杂、学习效率低，且易陷入局部最优解，影响诊断性能。针对上述问题，提出了一种基于改进型级联宽度学习（ICBL）的采煤机截割部齿轮箱故障诊断方法。在ICBL模型的特征节点中引入随机超图卷积机制，充分挖掘采煤机截割部齿轮箱振动数据的复杂多元结构信息，增强故障特征表征能力；采用类特异性权重分配策略，根据输入数据的类间比例信息，为少数类样本赋予更高权重，提高不平衡数据下采煤机截割部齿轮箱故障诊断性能。利用采煤机截割部齿轮箱故障模拟实验台验证基于ICBL的采煤机截割部齿轮箱故障诊断方法的有效性，结果表明该方法能够有效增强故障特征的判别性，在数据不平衡度为15时诊断精度达94.52%，单一样本的故障识别耗时为0.284 ms，优于级联宽度学习系统、加权宽度学习系统、多尺度卷积神经网络、超图神经网络、多分辨率超图卷积网络等。
- 采煤机截割部 /
- 齿轮箱 /
- 故障诊断 /
- 级联宽度学习 /
- 随机超图卷积 /
- 类特异性权重
Abstract:
The vibration monitoring data of the shearer cutting unit gearbox has a complex structure and is prone to class imbalance issues, leading to frequent false positives in traditional machine learning-based fault diagnosis methods. Meanwhile, deep learning-based approaches often suffer from complex model structures, low learning efficiency, and susceptibility to local optima, negatively impacting diagnostic performance. To address these issues, a fault diagnosis method was proposed for the shearer cutting unit gearbox based on improved cascaded broad learning (ICBL). A random hypergraph convolution mechanism was introduced into the feature nodes of the ICBL model to fully exploit the complex multivariate structural information in the vibration data of the shearer cutting unit gearbox, thereby enhancing the representation of fault features. Additionally, a class-specific weight allocation strategy was adopted to assign higher weights to minority class samples based on the class distribution of the input data, improving fault diagnosis performance under imbalanced data conditions. The effectiveness of the ICBL-based fault diagnosis method was validated using a shearer cutting unit gearbox fault simulation test platform. Experimental results demonstrated that the proposed method effectively enhanced the discriminability of fault features, achieving a diagnostic accuracy of 94.52% when the data imbalance ratio was 15, with a fault recognition time of 0.284 ms per sample. The method outperformed cascaded broad learning systems, weighted broad learning systems, multi-scale convolutional neural networks, hypergraph neural networks, and multi-resolution hypergraph convolutional networks, demonstrating significant engineering application value.
- shearer cutting unit /
- gearbox /
- fault diagnosis /
- cascaded broad learning /
- random hypergraph convolution /
- class-specific weight

HTML全文

图 1 CBLS模型结构

Figure 1. Structure of cascade broad learning system（CBLS） model

下载: 全尺寸图片幻灯片

图 2 ICBL模型结构

Figure 2. Structure of the ICBL model

下载: 全尺寸图片幻灯片

图 3 基于ICBL模型的采煤机截割部齿轮箱故障诊断流程

Figure 3. Fault diagnosis flow for shearer cutting unit gearbox based on the ICBL model

下载: 全尺寸图片幻灯片

图 4 采煤机截割部齿轮箱故障模拟实验台

Figure 4. Fault simulation test bench for shearer cutting unit gearbox

下载: 全尺寸图片幻灯片

图 5 各状态的振动信号及其Hilbert边际谱

Figure 5. Vibration signals and Hilbert marginal spectra for each state

下载: 全尺寸图片幻灯片

图 6 ICBL模型在数据集DA上的分类混淆矩阵

Figure 6. Classification confusion matrix of ICBL model on the dataset DA

下载: 全尺寸图片幻灯片

图 7 各模型在数据集DA上的10次实验结果

Figure 7. Results of 10 trials for each model on the dataset DA

下载: 全尺寸图片幻灯片

图 8 各模型在数据集DB—DF上的G−mean

Figure 8. G-mean of different models on the datasets DB-DF

下载: 全尺寸图片幻灯片

图 9 各模型在数据集DB—DF上的F1−score

Figure 9. F1-score of different models on the datasets DB-DF

下载: 全尺寸图片幻灯片

图 10 ICBL1和ICBL模型的特征可视化

Figure 10. Feature visualization of the ICBL1 and the ICBL model

下载: 全尺寸图片幻灯片

表 1 DA—DF数据集详细信息

Table 1 Detailed information of datasets DA-DF

数据集	训练样本个数					不平衡度
数据集	NS	MT	RC	GW	BT	不平衡度
DA	150	150	150	150	150	1
DB	150	75	75	75	75	2
DC	150	50	50	50	50	3
DD	150	25	25	25	25	6
DE	150	15	15	15	15	10
DF	150	10	10	10	10	15

下载: 导出CSV

表 2 不同输入方式下ICBL模型的故障诊断精度

Table 2 Fault diagnosis accuracy of ICBL under different inputs %

输入方式	不同数据集的故障诊断精度
输入方式	DA	DB	DC	DD	DE	DF
原始时域信号	92.76	93.59	90.61	90.56	88.40	86.08
频谱	94.55	94.65	93.10	94.31	93.10	93.10
包络谱	96.17	94.37	93.03	92.66	90.57	88.11
Hilbert边际谱	99.15	98.83	96.23	96.04	94.82	94.16

下载: 导出CSV

表 3 各模型在数据集DA上10次实验统计结果

Table 3 Statistical results of 10 trials for each model on the dataset DA

模型	平均精度/%	标准差/%	平均训练时间/s	平均诊断时间/s
CBLS	94.34	1.17	2.42	0.105
WBLS	92.19	1.68	3.51	0.123
MCNN	96.42	2.05	67.56	0.278
HGNN	98.04	0.71	90.57	0.319
MHNN	98.12	0.97	81.46	0.461
ICBL	99.54	0.28	4.72	0.142

下载: 导出CSV

表 4 不同噪声强度下各模型的故障诊断精度

Table 4 Fault diagnosis accuracy of each model under different noise intensity %

模型	不同噪声强度下故障诊断精度
模型	0	2 dB	4 dB	6 dB	8 dB
CBLS	81.06	86.64	89.46	93.88	95.14
WBLS	81.69	84.27	91.17	93.91	94.01
MCNN	84.79	89.51	91.13	91.80	93.30
HGNN	85.95	85.12	89.00	91.45	95.58
MHNN	88.09	91.37	90.00	91.97	94.70
ICBL	92.69	94.00	94.35	94.60	97.37

下载: 导出CSV

表 5 不同模型在数据集DB—DF上的故障诊断精度

Table 5 Fault diagnosis accuracy of different models on the datasets DB-DF %

模型	不同数据集上的故障诊断精度
模型	DB	DC	DD	DE	DF
CBLS	93.66	91.97	91.34	89.81	86.17
WBLS	92.23	92.77	91.92	93.02	89.59
MCNN	94.30	94.43	91.18	85.35	82.59
HGNN	95.85	93.05	90.55	87.11	86.12
MHNN	96.88	94.74	91.09	86.80	81.52
ICBL	98.53	96.42	96.09	94.89	94.52

下载: 导出CSV

表 6 在数据集DA—DF上的消融实验结果

Table 6 Results of ablation experiments on the datasets DA-DF %

模型	不同数据集上的故障诊断精度
模型	DA	DB	DC	DD	DE	DF
CBLS	94.34	93.66	91.97	91.34	89.81	86.17
ICBL1	96.47	96.35	94.62	95.62	93.16	93.29
ICBL2	97.48	96.24	94.52	93.00	91.12	89.76
ICBL	99.54	98.53	96.42	96.09	94.89	94.52

下载: 导出CSV

参考文献(24)

[1]	WANG Xiangrong,LI Congming,TIAN Hongying,et al. A step-by-step parameter-adaptive FMD method and its application in fault diagnosis[J]. Measurement Science and Technology,2024,35(4). DOI: 10.1088/1361-6501/AD197B.
[2]	尹玉玺,周常飞,史春祥,等. 采煤机摇臂齿轮箱故障诊断研究现状及展望[J]. 煤炭工程,2022,54(11):107-112. YIN Yuxi,ZHOU Changfei,SHI Chunxiang,et al. Research status and prospect of fault diagnosis for shearer rocker gear box[J]. Coal Engineering,2022,54(11):107-112.
[3]	LI Xin,LI Yong,YAN Ke,et al. An intelligent fault detection method of industrial gearboxes with robustness one-class support matrix machine toward multisource nonideal data[J]. IEEE/ASME Transactions on Mechatronics,2023,29(1):388-399.
[4]	刘旭南,赵丽娟,付东波,等. 采煤机截割部传动系统故障信号小波包分解方法研究[J]. 振动与冲击,2019,38(14):169-175,253. LIU Xunan,ZHAO Lijuan,FU Dongbo,et al. Study on wavelet packet decomposition method for fault signal of shearer cutting unit transmission system[J]. Journal of Vibration and Shock,2019,38(14):169-175,253.
[5]	张志刚,陈巧云,马俊. 基于小波包和EM聚类的采煤机齿轮故障诊断[J]. 煤矿机械,2020,41(9):183-186. ZHANG Zhigang,CHEN Qiaoyun,MA Jun. Fault diagnosis of shearer gear based on wavelet packet and EM clustering[J]. Coal Mine Machinery,2020,41(9):183-186.
[6]	LI Xin,LI Yong,YAN Ke,et al. Intelligent fault diagnosis of bevel gearboxes using semi-supervised probability support matrix machine and infrared imaging[J]. Reliability Engineering & System Safety,2023,230. DOI: 10.1016/J.RESS.2022.108921.
[7]	窦桂东,白艺硕,王均利,等. 基于深度学习的矿井滚动轴承故障诊断方法[J]. 工矿自动化,2024,50(1):96-103,154. DOU Guidong,BAI Yishuo,WANG Junli,et al. A fault diagnosis method for mine rolling bearings based on deep learning[J]. Journal of Mine Automation,2024,50(1):96-103,154.
[8]	包从望,江伟,刘永志,等. 基于卷积神经网络的采煤机截割部减速器故障诊断研究[J]. 机电工程,2021,38(10):1317-1323,1331. DOI: 10.3969/j.issn.1001-4551.2021.10.014 BAO Congwang,JIANG Wei,LIU Yongzhi,et al. Fault diagnosis method of shearer cutting unit reducer based on CNN[J]. Journal of Mechanical & Electrical Engineering,2021,38(10):1317-1323,1331. DOI: 10.3969/j.issn.1001-4551.2021.10.014
[9]	孙晓春,丁华,牛锐祥,等. 基于LW−DenseNet的采煤机摇臂齿轮故障诊断[J]. 煤炭工程,2023,55(11):186-192. SUN Xiaochun,DING Hua,NIU Ruixiang,et al. Fault diagnosis of rocker gear of shearer based on LW-DenseNet[J]. Coal Engineering,2023,55(11):186-192.
[10]	SONG Yang,WANG Weidong,WU Yuxin,et al. Unsupervised anomaly detection in shearers via autoencoder networks and multi-scale correlation matrix reconstruction[J]. International Journal of Coal Science & Technology,2024,11(1). DOI: 10.1007/S40789-024-00730-9.
[11]	邹筱瑜,孙国庆,王忠宾,等. 基于时频融合深度网络的矿用钻机轴承故障诊断[J]. 中国机械工程,2024,35(8):1405-1413,1448. DOI: 10.3969/j.issn.1004-132X.2024.08.009 ZOU Xiaoyu,SUN Guoqing,WANG Zhongbin,et al. Bearing fault diagnosis of mining drilling rig with time-frequency-fused deep network[J]. China Mechanical Engineering,2024,35(8):1405-1413,1448. DOI: 10.3969/j.issn.1004-132X.2024.08.009
[12]	毛君,郭浩,陈洪月. 基于深度自编码网络的采煤机截割部减速器故障诊断[J]. 煤炭科学技术,2019,47(11):123-128. MAO Jun,GUO Hao,CHEN Hongyue. Fault diagnosis of shearer cutting unit reducer based on deep auto-encoder network[J]. Coal Science and Technology,2019,47(11):123-128.
[13]	李勇,张启志,庄德玉,等. 强噪声干扰下采煤机行星齿轮故障诊断方法[J]. 工矿自动化,2024,50(6):142-149. LI Yong,ZHANG Qizhi,ZHUANG Deyu,et al. Diagnosis method for planetary gear faults in shearer under strong noise interference[J]. Journal of Mine Automation,2024,50(6):142-149.
[14]	CHEN C L P,LIU Zhulin,FENG Shuang. Universal approximation capability of broad learning system and its structural variations[J]. IEEE Transactions on Neural Networks and Learning Systems,2019,30(4):1191-1204. DOI: 10.1109/TNNLS.2018.2866622
[15]	YI Jun,HUANG Jiahua,ZHOU Wei,et al. Intergroup cascade broad learning system with optimized parameters for chaotic time series prediction[J]. IEEE Transactions on Artificial Intelligence,2022,3(5):709-721.
[16]	YE Hailiang,LI Hong,CHEN C L P. Adaptive deep cascade broad learning system and its application in image denoising[J]. IEEE Transactions on Cybernetics,2021,51(9):4450-4463. DOI: 10.1109/TCYB.2020.2978500
[17]	王召新,续欣莹,刘华平,等. 基于级联宽度学习的多模态材质识别[J]. 智能系统学报,2020,15(4):787-794. WANG Zhaoxin,XU Xinying,LIU Huaping,et al. Cascade broad learning for multi-modal material recognition[J]. CAAI Transactions on Intelligent Systems,2020,15(4):787-794.
[18]	ZHANG Kongliang,LI Hongkun,CAO Shunxin,et al. Motor current signal analysis using hypergraph neural networks for fault diagnosis of electromechanical system[J]. Measurement,2022,201. DOI: 10.1016/J.MEASUREMENT.2022.111697.
[19]	FENG Yifan,YOU Haoxuan,ZHANG Zizhao,et al. Hypergraph neural networks[C]. The AAAI Conference on Artificial Intelligence,Hawaii,2019:3558-3565.
[20]	CHU Fei,LIANG Tao,CHEN C L P,et al. Weighted broad learning system and its application in nonlinear industrial process modeling[J]. IEEE Transactions on Neural Networks and Learning Systems,2020,31(8):3017-3031. DOI: 10.1109/TNNLS.2019.2935033
[21]	GAO Jianhong,GUO Moufa,LIN Shuyue,et al. Application of semantic segmentation in high-impedance fault diagnosis combined signal envelope and Hilbert marginal spectrum for resonant distribution networks[J]. Expert Systems with Applications,2023,231. DOI: 10.1016/j.eswa.2023.
[22]	CAO Yudong,DING Yifei,JIA Minping,et al. A novel temporal convolutional network with residual self-attention mechanism for remaining useful life prediction of rolling bearings[J]. Reliability Engineering & System Safety,2021,215. DOI: 10.1016/J.RESS.2021.107813.
[23]	JIANG Guoqian,HE Haibo,YAN Jun,et al. Multiscale convolutional neural networks for fault diagnosis of wind turbine gearbox[J]. IEEE Transactions on Industrial Electronics,2018,66(4):3196-3207.
[24]	YAN Xunshi,LIU Yang,ZHANG Chen'an. Multiresolution hypergraph neural network for intelligent fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement,2022,71:1-10.

施引文献(13)

期刊类型引用(8)

1.	张玉涛，路旭，李亚清，张园勃，车博. 低甲烷气氛下褐煤的低温氧化特性研究. 安全与环境学报. 2024(02): 517-524 . 百度学术
2.	王斌，贾澎涛，郭风景，孙刘咏，林开义. 基于多特征融合的煤自燃温度深度预测模型. 中国矿业. 2024(02): 84-90 . 百度学术
3.	田富超，贾东旭，陈明义，梁运涛，朱红青，张同浩. 采空区复合灾害环境下含瓦斯煤自燃特征研究进展. 煤炭学报. 2024(06): 2711-2727 . 百度学术
4.	叶正亮，龚选平，尚博，胡冕. 煤自燃全过程精准预测预报指标体系研究与应用. 矿业研究与开发. 2023(09): 141-145 . 百度学术
5.	贾澎涛，林开义，郭风景. 基于PSO-SRU深度神经网络的煤自燃温度预测模型. 工矿自动化. 2022(04): 105-113 . 本站查看
6.	董轩萌，郭立稳，张少华，董宪伟，王福生. 氧化煤阻化性能的FTIR研究. 工业安全与环保. 2021(07): 34-39 . 百度学术
7.	骆大勇，刘振. 许疃煤矿煤炭自燃指标气体优选与应用. 矿业安全与环保. 2021(05): 69-74 . 百度学术
8.	刘宝，穆坤，叶飞，汪帆，王静婷. 基于相关向量机的煤自燃预测方法. 工矿自动化. 2020(09): 104-108 . 本站查看