Conjoint localization dense networks for fault feature extraction of variable load gearbox

doi:10.16265/j.cnki.issn1003-3033.2024.10.1296

Abstract

Abstract:

To address the challenge of extracting pulse signals in fault diagnosis of variable-load gearbox caused by redundant features, a pulse feature extraction method based on CAM was proposed. First, a CAM was designed, which consisted of two stages. In the first stage, a multilayer perceptron was used to simulate the channel dependencies and enhanced the important channel features related to faults. In the second stage, the convolutional layers were employed to learn signal segments related to faults. By recalibrating the features in two stages, the module focused on the critical pulse features. Next, based on CAM, this study proposed a CLDN method for extracting fault features in variable-load gearboxes. CLDN further improved the learning and representation of impulse signals by adaptively recalibrating the features at each layer. Finally, the extracted features were fed into a Softmax classifier to validate the feature extraction effect of the proposed method. The results show that CAM's accuracy is on average 3.8% higher than 4 attention mechanisms like Self-Attention, achieving accurate localization of impulse features. Compared with 7 diagnostic methods such as ResNet34, the accuracy of CLDN is 3.7% to 14.6% higher, which significantly enhances the extraction of fault features.

Key words: conjoint localization dense networks (CLDN), variable-load gearbox, fault diagnosis, feature extraction, conjoint attention module(CAM)

CLC Number:

X941

FAN Xiaoxuan, DUAN Lixiang, ZHANG Na, LI Xingtao, JIANG Lumeng. Conjoint localization dense networks for fault feature extraction of variable load gearbox[J]. China Safety Science Journal, 2024, 34(10): 166-173.

Figures/Tables 13

Fig.1

Fig.2

Fig.3

Fig.4

Table 1

Fig.5

Table 2

Fig.6

Table 3

Fig.7

Table 4

Fig.8

Fig.9

References 11

[1]	KUMAR S, KUMAR V, SARANGI S, et al. Gearbox fault diagnosis: a higher order moments approach[J]. Measurement, 2023, 210: DOI: 10.1016/j.measurement.2023.112489.
[2]	HUSSAIN S, GABBAR H. A novel method for real time gear fault detection based on pulse shape analysis[J]. Mechanical Systems and Signal Processing, 2011, 25(4): 1287-1298.
[3]	LI Jimeng, YAO Xifeng, WANG Xiangdong, et al. Multiscale local features learning based on BP neural network for rolling bearing intelligent fault diagnosis[J]. Measurement, 2020, 153: DOI: 10.1016/j.measurement.2019.107419.
[4]	张斌, 张辛忻, 杨海涛. 基于神经网络的振动给料机故障诊断算法[J]. 中国安全科学学报, 2023, 33(增2):55-59.
	ZHANG Bin, ZHANG Xinxin, YANG Haitao. Neural network-based fault diagnosis algorithm for vibration feeder[J]. China Safety Science Journal, 2023, 33(S2): 55-59. doi: 10.16265/j.cnki.issn1003-3033.2023.S2.0019
[5]	王庆荣, 王媛, 朱昌锋, 等. 双通道交叉密集连接的滚动轴承故障诊断[J/OL]. 机械科学与技术: 1-9 [2023-04-06]. https://doi.org/10.13433/j.cnki.1003-8728.20230188.
	WANG Qingrong, WANG Yuan, ZHU Changfeng, et al. Fault diagnosis of rolling bearings with two-channel cross-dense connection[J]. Mechanical Science and Technology for Aerospace Engineering: 1-9 [2023-04-06]. https://doi.org/10.13433/j.cnki.1003-8728.20230188.
[6]	WOO S, PARK J, LEE J, et al. CBAM: convolutional block attention module[C]. Proceedings of the 15^th European Conference on Computer Vision (ECCV), 2018: 3-19.
[7]	HU Jie, SHEN Li, SUN Gang. Squeeze-and-excitation networks[C]. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2018: 7132-7141.
[8]	TANG Shenggan, ZHU Yong, YUAN Shouqi. Intelligent fault diagnosis of hydraulic piston pump based on deep learning and Bayesian optimization[J]. ISA Transactions, 2022, 129: 555-563. doi: 10.1016/j.isatra.2022.01.013 pmid: 35115164
[9]	ZHAO Xiaoli, YAO Jianyong, DENG Wenxiang, et al. Multiscale deep graph convolutional networks for intelligent fault diagnosis of rotor-bearing system under fluctuating working conditions[J]. IEEE Transactions on Industrial Informatics, 2023, 19(1): 166-176.
[10]	王金江, 关鹏婷, 陈卓, 等. 基于深度学习的检修作业过程风险智能预警[J]. 中国安全科学学报, 2023, 33(10): 16-22. doi: 10.16265/j.cnki.issn1003-3033.2023.10.1744
	WANG Jinjiang, GUAN Pengting, CHEN Zhuo, et al. Intelligent warning of risk during maintenance operations based on deep learning[J]. China Safety Science Journal, 2023, 33(10): 16-22. doi: 10.16265/j.cnki.issn1003-3033.2023.10.1744
[11]	MENG Zong, SHI Guixia, WANG Fulin. Vibration response and fault characteristics analysis of gear based on time-varying mesh stiffness[J]. Mechanism and Machine Theory, 2020, 148: DOI: 10.1016/j.mechmachtheory.2020.103786.

网络	各类故障状态诊断精度					总体诊断精度
网络	正常	剥落	点蚀	断齿	缺齿	总体诊断精度
DenseNet	92.22	100.00	96.67	88.89	100.00	95.56
DenseNet-FAM	97.78	100.00	96.67	85.56	100.00	96.00
DenseNet-SAM	94.44	100.00	96.67	97.78	100.00	97.78
CLDN	100.00	95.56	100.00	98.89	100.00	98.89

信噪比/dB	各网络诊断精度
信噪比/dB	CNN	ResNet34	DenseNet	Self-Attention	CBAM	ECA	No-local	CLDN
1	89.11	96	97.11	94.89	96.67	98.22	97.78	98.67
0	87.33	94.89	96.67	92.89	92.44	98	94.22	98.44
-1	75.78	94.67	95.56	91.11	92.45	96.67	92.89	98.89
平均值	84.07	95.19	96.45	92.96	93.85	97.63	94.96	98.67

故障类型	负载/kW×转速/ (r·min^-1)	训练	验证	测试	故障标签
正常	2.250×1 730	140	30	30	0
内圈故障IR1	0×1 797	140	30	30	1
内圈故障IR2		140	30	30	2
内圈故障IR3		140	30	30	3
外圈故障OR1	0.735×1 772	140	30	30	4
外圈故障OR2		140	30	30	5
外圈故障OR3		140	30	30	6
滚动体故障BA1	1.470×1 750	140	30	30	7
滚动体故障BA2		140	30	30	8
滚动体故障BA3		140	30	30	9

[1]	WANG Yantao, GAO Yi, SHI Tongyu. A review of flight control system fault research based on QAR data [J]. China Safety Science Journal, 2024, 34(4): 1-9.
[2]	SHANG Shuhong, CAO Liang, LU Wei. Fault diagnosis algorithm for medium draining screen based on wavelet transform feature extraction [J]. China Safety Science Journal, 2023, 33(S2): 233-237.
[3]	ZHANG Bin, ZHANG Xinxin, YANG Haitao. Neural network-based fault diagnosis algorithm for vibration feeder [J]. China Safety Science Journal, 2023, 33(S2): 55-59.
[4]	LIU Xin, WANG Yajun, YAO Meichang, SU Lei, WANG Peng. Online monitoring of hydraulic brake of mine hoist based on LabVIEW [J]. China Safety Science Journal, 2023, 33(S1): 145-149.
[5]	ZHANG Qi, JIRGEL. Detection algorithm of slime water overflow based on convolutional neural network [J]. China Safety Science Journal, 2023, 33(S1): 192-195.
[6]	YIN Yibing, ZHANG Yongliang, LIU Zunmin, ZHANG Qiang. Detection method and experiment of small hole leakage based on electrostatic sensing [J]. China Safety Science Journal, 2023, 33(9): 142-149.
[7]	LI Shanmei, ZHOU Xiangzhi. Prediction of airport departure delay based on S2S-CNN-GRU [J]. China Safety Science Journal, 2023, 33(8): 93-100.
[8]	LU Caiwu, GAO Ruiyang, XU Xiaohui, JIANG Song, LIU Di, FU Xinkai. Study on miners' attention under different noise conditions based on EEG data [J]. China Safety Science Journal, 2023, 33(5): 35-41.
[9]	YANG Yu, YANG Xin, WANG Ying, ZHAI Chi, ZHANG Hao. Research on TE process fault diagnosis based on mini-1D-CNN model [J]. China Safety Science Journal, 2023, 33(2): 173-178.
[10]	PAN Weijun, QIN Liru, LUO Chen, HUANG Yuanjing. Adaptive feature extraction of general aviation forest fire rescue data [J]. China Safety Science Journal, 2023, 33(11): 67-74.
[11]	YU Wei, WANG Zhipeng, LIU Shuo, HE Gang. Inter-turn short circuit fault diagnosis for motors based on multi-level fine model [J]. China Safety Science Journal, 2022, 32(6): 95-102.
[12]	LIN Haixiang, LU Ran, LU Renjie, LI Xinqin, ZHAO Zhengxiang, BAI Wansheng. Fault diagnosis of high-speed railway on-board equipment based on BiLSTM-CBA hybrid model [J]. China Safety Science Journal, 2022, 32(6): 79-86.
[13]	LI Gang, LU Peiling, YANG Yong. Research on application of ZPW-2000 fault diagnosis algorithm for track circuits [J]. China Safety Science Journal, 2022, 32(6): 53-59.
[14]	XING Yulong, WANG Jian, SHANGGUAN Wei, PENG Cong, ZHU Linfu. Track circuit fault diagnosis method for massive imbalanced data [J]. China Safety Science Journal, 2022, 32(5): 112-118.
[15]	ZHAO Dan, SHEN Zhiyuan, LIU Xiaoqing. Online fault diagnosis of mine ventilation system based on OCISVM [J]. China Safety Science Journal, 2022, 32(10): 76-82.