Abstract
Marine power-generation diesel engines operate in harsh environments. Their vibration signals are highly complex and the feature information exhibits a non-linear distribution. It is difficult to extract effective feature information from the network model, resulting in low fault-diagnosis accuracy. To address this problem, we propose a fault-diagnosis method that combines the Gramian angular field (GAF) with a convolutional neural network (CNN). Firstly, the vibration signals are transformed into 2D images by taking advantage of the GAF, which preserves the temporal correlation. The raw signals can be mapped to 2D image features such as texture and color. To integrate the feature information, the images of the Gramian angular summation field (GASF) and Gramian angular difference field (GADF) are fused by the weighted average fusion method. Secondly, the channel attention mechanism and temporal attention mechanism are introduced in the CNN model to optimize the CNN learning mechanism. Introducing the concept of residuals in the attention mechanism improves the feasibility of optimization. Finally, the weighted average fused images are fed into the CNN for feature extraction and fault diagnosis. The validity of the proposed method is verified by experiments with abnormal valve clearance. The average diagnostic accuracy is 98.40%. When −20 dB⩽signal-to-noise ratio (SNR)⩽20 dB, the diagnostic accuracy of the proposed method is higher than 94.00%. The proposed method has superior diagnostic performance. Moreover, it has a certain anti-noise capability and variable-load adaptive capability.
概要
目的
船用发电柴油机工作环境恶劣, 在内外多激励源的干扰下, 振动信号呈现非线性非平稳性的特点。本文旨在对船舶发电柴油机的振动信号进行有效特征提取并准确识别故障类型。同时, 研究所提方法的有效性, 以提高船舶发电柴油机的故障诊断精度。
创新点
1. 一维振动信号可通过格拉姆角场转换为二维图像;一维振动信号可以映射到二维图像的颜色、点、线和其他特征;为充分利用故障特征信息, 将格拉姆角和场和格拉姆角差场获得的图像进行加权平均融合。2. 利用多注意力机制来优化卷积神经网络学习机制, 使网络有选择地提取信号中的关键特征信息。
方法
1. 对船舶发电柴油机的气阀间隙进行故障预设, 采集柴油机不同健康状态振动信号。2. 将振动信号转化为二维图像, 并将格拉姆角和场和格拉姆角差场获得的图像进行加权平均融合, 以充分利用原信号中的故障特征信息。3. 将融合后的图像输入到卷积神经网络中进行自适应特征提取和故障识别。
结论
1. 所提方法可准确识别故障类型,平均诊断精度可达98.40%;与其他方法相比,所提方法具有更高的故障诊断精度。2. 在不同信噪比下, 所提方法与无注意力机制方法相比, 准确精度可提高14.80%。3. 融合后的图像可为神经网络提供更充足的特征信息,因此具有更高的故障识别精度。4. 变负荷实验中, 所提方法的准确率均保持在89.00%以上, 进一步验证了所提方法的稳定性。
Similar content being viewed by others
References
Alsalaet JK, Hajnayeb A, Bahedh AS, 2023. Bearing fault diagnosis using normalized diagnostic feature-gram and convolutional neural network. Measurement Science and Technology, 34(4):045901. https://doi.org/10.1088/1361-6501/ACAD1F
Cai BP, Sun XT, Wang JX, et al., 2020. Fault detection and diagnostic method of diesel engine by combining rule-based algorithm and BNs/BPNNs. Journal of Manufacturing Systems, 57:148–157. https://doi.org/10.1016/j.jmsy.2020.09.001
Cerrada M, Zurita G, Cabrera D, et al., 2016. Fault diagnosis in spur gears based on genetic algorithm and random forest. Mechanical Systems and Signal Processing, 70–71: 87–103. https://doi.org/10.1016/j.ymssp.2015.08.030
Cui JL, Zhong QW, Zheng SB, et al., 2022. A lightweight model for bearing fault diagnosis based on Gramian angular field and coordinate attention. Machines, 10(4):282. https://doi.org/10.3390/MACHINES10040282
Dhamande LS, Chaudhari MB, 2016. Bearing fault diagnosis based on statistical feature extraction in time and frequency domain and neural network. International Journal of Vehicle Structures and Systems, 8(4):229–240. https://doi.org/10.4273/ijvss.8.4.09
Du JF, Li XY, Gao YP, et al., 2022. Integrated gradient-based continuous wavelet transform for bearing fault diagnosis. Sensors, 22(22):8760. https://doi.org/10.3390/S22228760
Fu WL, Jiang XH, Li BL, et al., 2023. Rolling bearing fault diagnosis based on 2D time-frequency images and data augmentation technique. Measurement Science and Technology, 34(4):045005. https://doi.org/10.1088/1361-6501/ACABDB
Gou LF, Li HH, Zheng H, et al., 2020. Aeroengine control system sensor fault diagnosis based on CWT and CNN. Mathematical Problems in Engineering, 2020:5357146. https://doi.org/10.1155/2020/5357146
Grover C, Turk N, 2022. A novel fault diagnostic system for rolling element bearings using deep transfer learning on bispectrum contour maps. Engineering Science and Technology, an International Journal, 31:101049. https://doi.org/10.1016/J.JESTCH.2021.08.006
Hajnayeb A, 2021. Cavitation analysis in centrifugal pumps based on vibration bispectrum and transfer learning. Shock and Vibration, 2021:6988949. https://doi.org/10.1155/2021/6988949
He Y, Tang HS, Ren Y, et al., 2022. A deep multi-signal fusion adversarial model based transfer learning and residual network for axial piston pump fault diagnosis. Measurement, 192:110889. https://doi.org/10.1016/J.MEASUREMENT.2022.110889
Hoang DT, Kang HJ, 2019. A survey on deep learning based bearing fault diagnosis. Neurocomputing, 335:327–335. https://doi.org/10.1016/j.neucom.2018.06.078
Hou SZ, Guo W, Wang ZQ, et al., 2022. Deep-learning-based fault type identification using modified CEEMDAN and image augmentation in distribution power grid. IEEE Sensors Journal, 22(2):1583–1596. https://doi.org/10.1109/JSEN.2021.3133352
Hu J, Yu YH, Yang JG, et al., 2023. Research on the generalisation method of diesel engine exhaust valve leakage fault diagnosis based on acoustic emission. Measurement, 210:112560. https://doi.org/10.1016/J.MEASUREMENT.2023.112560
Karatuğ Ç, Arslanoğlu Y, 2022. Development of condition-based maintenance strategy for fault diagnosis for ship engine systems. Ocean Engineering, 256:111515. https://doi.org/10.1016/J.OCEANENG.2022.111515
Manarikkal I, Elasha F, Mba D, 2021. Diagnostics and prognostics of planetary gearbox using CWT, auto regression (AR) and K-means algorithm. Applied Acoustics, 184: 108314. https://doi.org/10.1016/J.APACOUST.2021.108314
Nayana BR, Geethanjali P, 2017. Analysis of statistical timedomain features effectiveness in identification of bearing faults from vibration signal. IEEE Sensors Journal, 17(17): 5618–5625. https://doi.org/10.1109/jsen.2017.2727638
Pan JH, Qu LL, Peng KX, 2021. Sensor and actuator fault diagnosis for robot joint based on deep CNN. Entropy, 23(6): 751. https://doi.org/10.3390/E23060751
Peng DD, Wang H, Liu ZL, et al., 2020. Multibranch and multiscale CNN for fault diagnosis of wheelset bearings under strong noise and variable load condition. IEEE Transactions on Industrial Informatics, 16(7):4949–4960. https://doi.org/10.1109/tii.2020.2967557
Qian CH, Zhu JJ, Shen YH, et al., 2022. Deep transfer learning in mechanical intelligent fault diagnosis: application and challenge. Neural Processing Letters, 54(3):2509–2531. https://doi.org/10.1007/S11063-021-10719-Z
Rao X, Sheng CX, Guo ZW, et al., 2022. A review of online condition monitoring and maintenance strategy for cylinder liner-piston rings of diesel engines. Mechanical Systems and Signal Processing, 165:108385. https://doi.org/10.1016/J.YMSSP.2021.108385
Ren HR, Liao XJ, Li ZW, et al., 2018. Anomaly detection using piecewise aggregate approximation in the amplitude domain. Applied Intelligence, 48(5):1097–1110. https://doi.org/10.1007/s10489-017-1017-x
Ren K, Zhang DW, Wan MJ, et al., 2021. An infrared and visible image fusion method based on improved DenseNet and mRMR-ZCA. Infrared Physics & Technology, 115:103707. https://doi.org/10.1016/J.INFRARED.2021.103707
Schmidhuber J, 2015. Deep learning in neural networks: an overview. Neural Networks, 61:85–117. https://doi.org/10.1016/j.neunet.2014.09.003
Senanayaka JSL, van Khang H, Robbersmyr KG, 2019. Multiple classifiers and data fusion for robust diagnosis of gearbox mixed fault. IEEE Transactions on Industrial Informatics, 15(8):4569–4579. https://doi.org/10.1109/TII.2018.2883357
Song RW, Yu BQ, Shi H, et al., 2023. Support vector machine fault diagnosis based on sparse scaling convex hull. Measurement Science and Technology, 34(3):035101. https://doi.org/10.1088/1361-6501/ACA217
Sun F, Xu H, Zhao YH, et al., 2022. Data-driven fault diagnosis of control valve with missing data based on modeling and deep residual shrinkage network. Journal of Zhejiang University-SCIENCE A (Applied Physics & Engineering), 23(4):303–313. https://doi.org/10.1631/jzus.A2100598
Tian HX, Li RJ, Yang LZ, 2022. Operation status monitoring of reciprocating compressors based on the fusion of spatiotemporal multiple information. Measurement, 204:112087. https://doi.org/10.1016/J.MEASUREMENT.2022.112087
Wang B, Lei YG, Li NP, et al., 2021. Multiscale convolutional attention network for predicting remaining useful life of machinery. IEEE Transactions on Industrial Electronics, 68(8):7496–7504. https://doi.org/10.1109/tie.2020.3003649
Wen L, Li XY, Gao L, 2020. A transfer convolutional neural network for fault diagnosis based on ResNet-50. Neural Computing and Applications, 32(10):6111–6124. https://doi.org/10.1007/s00521-019-04097-w
Xie JS, Lin MQ, Yang BY, et al., 2023. A novel bearing fault diagnosis method under small samples using time-frequency multi-scale convolution layer and hybrid attention mechanism module. Measurement Science and Technology, 34(9):095121. https://doi.org/10.1088/1361-6501/ACDC45
Xu YD, Yan XA, Feng K, et al., 2022. Attention-based multiscale denoising residual convolutional neural networks for fault diagnosis of rotating machinery. Reliability Engineering & System Safety, 226:108714. https://doi.org/10.1016/J.RESS.2022.108714
Zhao R, Yan RQ, Chen ZH, et al., 2019. Deep learning and its applications to machine health monitoring. Mechanical Systems and Signal Processing, 115:213–237. https://doi.org/10.1016/j.ymssp.2018.05.050
Acknowledgments
This work is supported by the Project of Shanghai Engineering Research Center for Intelligent Operation and Maintenance and Energy Efficiency Monitoring of Ships (No. 20DZ2252300), China. We sincerely thank Dexin CUI (the superintendent of Ningbo Ocean Shipping Co., Ltd., China) for providing experimental guidance.
Author information
Authors and Affiliations
Contributions
Congyue LI and Yihuai HU designed the research. Congyue LI wrote the first draft of the manuscript. Jiawei JIANG and Dexin CUI helped to organize the manuscript. Congyue LI and Yihuai HU revised and edited the final version.
Corresponding author
Ethics declarations
Congyue LI, Yihuai HU, Jiawei JIANG, and Dexin CUI declare that they have no conflict of interest.
Rights and permissions
About this article
Cite this article
Li, C., Hu, Y., Jiang, J. et al. Fault diagnosis of a marine power-generation diesel engine based on the Gramian angular field and a convolutional neural network. J. Zhejiang Univ. Sci. A (2024). https://doi.org/10.1631/jzus.A2300273
Received:
Accepted:
Published:
DOI: https://doi.org/10.1631/jzus.A2300273
Key words
- Multi-attention mechanisms (MAM)
- Convolutional neural network (CNN)
- Gramian angular field (GAF)
- Image fusion
- Marine power-generation diesel engine
- Fault diagnosis