矿井通风机是煤矿企业生产中极为重要的设备，轴承是维持通风机稳定运行的关键部件之一。但由于矿山工作环境的复杂性，矿井通风机轴承容易受到损伤故障。在运行过程中，通风机轴承故障可能会对整个通风机系统产生影响。由于机械部件如轴承在运行时经常会产生规律性的振动，因此对振动信号进行分析和处理对于监测设备状态和诊断故障非常重要。随着神经网络的发展，深度学习展示出卓越的特征提取和表示能力，充分利用大量数据自动学习轴承故障的特征信息，能够在复杂的情况下实现轴承故障诊断。因此，本研究以滚动轴承的振动信号为研究对象，将高噪声、不同负载等复杂场景下的端到端轴承故障诊断作为应用背景，探讨了不同深度学习神经网络在这些复杂环境下的应用。主要的研究内容如下：

（1）考虑到深度卷积神经网络有着强大的特征提取能力，能够替代手动挑选特征的过程，提出了端到端的轴承故障种类诊断模型。并利用凯斯西储大学轴承数据集对模型的故障诊断效果进行验证，在信噪比为10的环境下，故障诊断率接近100%。

（2）为了进一步提升高噪声环境下的轴承故障诊断性能，使用残差网络模块优化卷积神经网络，加大网络结构的深度，并使用大量稀疏化结构降低了网络的过拟合风险。并通过大量实验选择合适网络的超参数，提高了高噪声环境下矿井通风机滚动轴承故障诊断的准确率，相较于卷积神经网络在信噪比为0时的故障诊断率，提高了25个百分点。

（3）提出采用 Transformer编码器的ViT模型用于通风机故障诊断，使用增广数据进行训练后，ViT模型在高噪声环境下也有很高的故障诊断准确率，相较于残差神经网络在信噪比为-10高噪声环境时的故障诊断率，提高了10个百分点。并进行了可变负载情景下的轴承故障诊断，证明了在大数据上进行预训练，再用少量数据进行微调的训练方式可以用于矿井通风机故障诊断，为矿井通风机轴承故障诊断方式提供了新的思路，为未来进一步实现矿井通风机故障诊断智能化奠定了基础。还利用t-SNE网络可视化技术展示了神经网络最后的特征提取结果，提升神经网络可解释性。

Mine ventilators are extremely important equipment in the production of coal mining enterprises, and bearings are one of the key components to maintain the stable operation of ventilators. However, due to the complexity of the mine working environment, mine fan bearings are vulnerable to damage and failure. During operation, fan bearing failure may have an impact on the entire fan system. Since mechanical components such as bearings often vibrate regularly during operation, analyzing and processing vibration signals is very important for monitoring equipment status and diagnosing faults. With the development of neural networks, deep learning has demonstrated excellent feature extraction and representation capabilities, making full use of a large amount of data to automatically learn the characteristic information of bearing faults, and can realize bearing fault diagnosis in complex situations. Therefore, this study takes the vibration signals of rolling bearings as the research object, takes the end-to-end bearing fault diagnosis under complex scenarios such as high noise and different loads as the application background, and discusses the application of different deep neural networks in these complex environments. The main research content is as follows:

(1) Considering that the deep convolutional neural network has a powerful feature extraction ability and can replace the process of manually selecting features, an end-to-end bearing fault diagnosis model is proposed. The bearing data set of Case Western Reserve University is used to verify the fault diagnosis effect of the model. In the environment where the signal-to-noise ratio is 10, the fault diagnosis rate is close to 100%.

(2) In order to further improve the performance of bearing fault diagnosis in high-noise environments, the residual network module is used to optimize the convolutional neural network, the depth of the network structure is increased, and a large number of sparse structures are used to reduce the risk of over-fitting of the network. And through a large number of experiments to select the appropriate hyperparameters of the network, the accuracy of the fault diagnosis of the rolling bearing of the mine fan in the high-noise environment has been improved. percentage point.

(3) The ViT model using the Transformer encoder is proposed for fan fault diagnosis. After training with augmented data, the ViT model also has a high fault diagnosis accuracy in high-noise environments. Compared with the residual neural network in the signal-to-noise ratio The fault diagnosis rate in a high-noise environment is -10, which has increased by 10 percentage points. And the bearing fault diagnosis under the variable load scenario is carried out, which proves that the training method of pre-training on big data and then fine-tuning with a small amount of data can be used for fault diagnosis of mine ventilator, and provides a basis for mine ventilator bearing fault diagnosis. A new idea was established, which laid a foundation for the further realization of intelligent fault diagnosis of mine ventilators in the future. The t-SNE network visualization technology is also used to display the final feature extraction results of the neural network to improve the interpretability of the neural network.

康红普,王国法.煤炭行业高质量发展研究[J].中国工程科学,2021,23(05):130-138.

国家发展改革委.能源技术革命创新行动计划（2016-2030年）[EB/OL]. [2016-4-17].https://www.ndrc.gov.cn/xxgk/zcfb/tz/W020190905517012835441.pdf.

王国法.“碳中和”不是零碳，“碳达峰”也绝不是能源达峰[J].山西煤炭,2021,41(03):1-2.

王国法,张德生.煤炭智能化综采技术创新实践与发展展望[J].中国矿业大学学报,2018,47(03):459-467.

张国枢.通风安全学（第三版）[M].徐州:中国矿业大学出版社,2021.

雷亚国,贾峰.大数据下机械智能故障诊断的机遇与挑战[J].机械工程学报,2018,54(05):94-104.

何正嘉,陈进,王太勇,等.机械故障诊断理论及应用[M].高等教育出版社,2010.

Medearis K. Vibration analysis of rotating auxiliary machinery. Final report.[FAN computer code][R]. Medearis (Kenneth) and Associates, Fort Collins, CO (USA), 1978.

Piety K R, Pardue E F, Phillips E P. Experience With Periodic Vibration Monitoring In Fossil Power Stations[C]//Proceedings of the 16th Turbomachinery Symposium. Texas A&M University. Turbomachinery Laboratories, 1987.

Cory W T W. Overview of condition monitoring methods with emphasis on industrial fans[J]. Proceedings of the Institution of Mechanical Engineers, Part A: Journal of Power and Energy, 1991, 205(4): 225-240.

陈士玮,胡亚非.KJZ—2型矿井主通风机在线监测系统[J].流体机械,2000(04):40-42+4.

徐红波,陈国华,王新华.基于ARMA模型双谱分布与FCM方法的轴承故障识别[J].华南理工大学学报(自然科学版),2012,40(07):78-82+89.

Han T, Jiang D. Rolling bearing fault diagnostic method based on VMD-AR model and random forest classifier[J]. Shock and Vibration, 2016, 2016.

Leite V C M N, da Silva J G B, Veloso G F C, et al. Detection of localized bearing faults in induction machines by spectral kurtosis and envelope analysis of stator current[J]. IEEE Transactions on Industrial Electronics, 2014, 62(3): 1855-1865.

熊国良,胡俊锋,陈慧.基于SK-NLM包络的滚动轴承故障冲击特征增强[J].仪器仪表学报,2016,37(10):2176-2184.

Bhakta K, Sikder N, Al Nahid A, et al. Fault diagnosis of induction motor bearing using cepstrum-based preprocessing and ensemble learning algorithm[C]//2019 International Conf erence on Electrical, Computer and Communication Engineering (ECCE). IEEE, 2019: 1-6.

Cabal-Yepez E, Garcia-Ramirez A G, Romero-Troncoso R J, et al. Reconfigurable monitoring system for time-frequency analysis on industrial equipment through STFT and DWT[J]. IEEE transactions on industrial informatics, 2012, 9(2): 760-771.

Wang X, He Q. Machinery Fault Signal Reconstruction Using Time-Frequency Manifold[M]//Engineering Asset Management-Systems, Professional Practices and Certification. Springer, Cham, 2015: 777-787.

Yan R, Gao R X. Hilbert–Huang transform-based vibration signal analysis for machine health monitoring[J]. IEEE Transactions on Instrumentation and measurement, 2006, 55(6): 2320-2329.

Luo C, Jia M, Wen Y. The diagnosis approach for rolling bearing fault based on Kurtosis criterion EMD and Hilbert envelope spectrum[C]//2017 IEEE 3rd Information Technology and Mechatronics Engineering Conference (ITOEC). IEEE, 2017: 692-696.

Ma J, Wu J, Fan Y, et al. The rolling bearing fault feature extraction based on the LMD and envelope demodulation[J]. Mathematical Problems in Engineering, 2015, 2015.

王晓龙,唐贵基.一种基于连续小波变换的滚动轴承早期故障诊断新方法[J].推进技术,2016,37(08):1431-1437.

Khanam S, Tandon N, Dutt J K. Fault size estimation in the outer race of ball bearing using discrete wavelet transform of the vibration signal[J]. Procedia Technology, 2014, 14: 12-19.

Li N, Zhou R, Zhao X Z. Mechanical faulty signal denoising using a redundant non-linear second-generation wavelet transform[J]. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2011, 225(4): 799-808.

Wong M L D, Jack L B, Nandi A K. Modified self-organising map for automated novelty detection applied to vibration signal monitoring[J]. Mechanical systems and signal processing, 2006, 20(3): 593-610.

Jung U, Koh B H. Wavelet energy-based visualization and classification of high-dimensional signal for bearing fault detection[J]. Knowledge and Information Systems, 2015, 44: 197-215.

Li F, Wang J, Chyu M K, et al. Weak fault diagnosis of rotating machinery based on feature reduction with Supervised Orthogonal Local Fisher Discriminant Analysis[J]. Neurocomputing, 2015, 168: 505-519.

Dong S, Xu X, Chen R. Application of fuzzy C-means method and classification model of optimized K-nearest neighbor for fault diagnosis of bearing[J]. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2016, 38: 2255-2263.

He D, Li R, Zhu J. Plastic bearing fault diagnosis based on a two-step data mining approach[J]. IEEE Transactions on Industrial Electronics, 2012, 60(8): 3429-3440.

Li N, Zhou R, Zhao X Z. Mechanical faulty signal denoising using a redundant non-linear second-generation wavelet transform[J]. Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science, 2011, 225(4): 799-808.

Cheng Y, Yuan H, Liu H, et al. Fault diagnosis for rolling bearing based on SIFT-KPCA and SVM[J]. Engineering Computations, 2017.

Li P. Fault Diagnosis of Motor Rolling Bearing Based on GWO-SVM[J]. International Core Journal of Engineering, 2019, 5(10): 238-245.

王帅星,黄茜,王晓笋.WPT、PCA与SVM结合的滚动轴承故障程度诊断[J].机械设计与制造,2022(04):5-9.

Razzak M I, Naz S, Zaib A. Deep learning for medical image processing: Overview, challenges and the future[J]. Classification in BioApps: Automation of Decision Making, 2018: 323-350.

Kamath U, Liu J, Whitaker J. Deep learning for NLP and speech recognition[M]. Cham, Switzerland: Springer, 2019.

Bourlard H, Kamp Y. Auto-association by multilayer perceptrons and singular value decomposition[J]. Biological cybernetics, 1988, 59(4): 291-294.

He X, Wang D, Li Y, et al. A novel bearing fault diagnosis method based on gaussian restricted boltzmann machine[J]. Mathematical Problems in Engineering, 2016, 2016.

Shao H, Jiang H, Zhang X, et al. Rolling bearing fault diagnosis using an optimization deep belief network[J]. Measurement Science and Technology, 2015, 26(11): 115002.

LeCun Y, Bottou L, Bengio Y, et al. Gradient-based learning applied to document recognition[J]. Proceedings of the IEEE, 1998, 86(11): 2278-2324.

Hopfield J J. Neural networks and physical systems with emergent collective computational abilities[J]. Proceedings of the national academy of sciences, 1982, 79(8): 2554-2558.

Rumelhart D E, Hinton G E, Williams R J. Learning representations by back-propagating errors[J]. nature, 1986, 323(6088): 533-536.

Hochreiter S, Schmidhuber J. Long short-term memory[J]. Neural computation, 1997, 9(8): 1735-1780.

Cho K, Van Merriënboer B, Gulcehre C, et al. Learning phrase representations using RNN encoder-decoder for statistical machine translation[J]. arXiv preprint arXiv:1406.1078, 2014.

Wen T H, Gasic M, Mrksic N, et al. Semantically conditioned lstm-based natural language generation for spoken dialogue systems[J]. arXiv preprint arXiv:1508.01745, 2015.

Graves A, Mohamed A, Hinton G. Speech recognition with deep recurrent neural networks[C]//2013 IEEE international conference on acoustics, speech and signal processing. Ieee, 2013: 6645-6649.

Xie Y, Zhang T. Feature extraction based on DWT and CNN for rotating machinery fault diagnosis[C]//2017 29th Chinese Control And Decision Conference (CCDC). IEEE, 2017: 3861-3866.

Ding X, He Q. Energy-fluctuated multiscale feature learning with deep convnet for intelligent spindle bearing fault diagnosis[J]. IEEE Transactions on Instrumentation and Measurement, 2017, 66(8): 1926-1935.

Verstraete D, Ferrada A, Droguett E L, et al. Deep learning enabled fault diagnosis using time-frequency image analysis of rolling element bearings[J]. Shock and Vibration, 2017, 2017.

Li X, Zhang W, Ding Q. Cross-domain fault diagnosis of rolling element bearings using deep generative neural networks[J]. IEEE Transactions on Industrial Electronics, 2018, 66(7): 5525-5534.

杜小磊,陈志刚,许旭.基于小波卷积自编码器和LSTM网络的轴承故障诊断研究[J].机电工程,2019,36(07):663-668.

Zhao H, Yang X, Chen B, et al. Bearing fault diagnosis using transfer learning and optimized deep belief network[J]. Measurement Science and Technology, 2022, 33(6): 065009.

Li X, Zhang W, Ding Q, et al. Intelligent rotating machinery fault diagnosis based on deep learning using data augmentation[J]. Journal of Intelligent Manufacturing, 2020, 31(2): 433-452.

Chen Y, Peng G, Xie C, et al. ACDIN: Bridging the gap between artificial and real bearing damages for bearing fault diagnosis[J]. Neurocomputing, 2018, 294: 61-71.

Xia M, Li T, Xu L, et al. Fault diagnosis for rotating machinery using multiple sensors and convolutional neural networks[J]. IEEE/ASME transactions on mechatronics, 2017, 23(1): 101-110.

Zhang W, Peng G, Li C. Bearings fault diagnosis based on convolutional neural networks with 2-D representation of vibration signals as input[C]//MATEC web of conferences. EDP Sciences, 2017, 95: 13001.

Pan H, He X, Tang S, et al. An Improved Bearing Fault Diagnosis Method using One-Dimensional CNN and LSTM[J]. Strojniski Vestnik/Journal of Mechanical Engineering, 2018, 64.

Zhao R, Yan R, Wang J, et al. Learning to monitor machine health with convolutional bi-directional LSTM networks[J]. Sensors, 2017, 17(2): 273.

Yu L, Qu J, Gao F, et al. A novel hierarchical algorithm for bearing fault diagnosis based on stacked LSTM[J]. Shock and Vibration, 2019, 2019.

Lu C, Wang Z, Zhou B. Intelligent fault diagnosis of rolling bearing using hierarchical convolutional network based health state classification[J]. Advanced Engineering Informatics, 2017, 32: 139-151.

Li X, Zhang W, Ding Q. A robust intelligent fault diagnosis method for rolling element bearings based on deep distance metric learning[J]. Neurocomputing, 2018, 310: 77-95.

Liu H, Zhou J, Zheng Y, et al. Fault diagnosis of rolling bearings with recurrent neural network-based autoencoders[J]. ISA transactions, 2018, 77: 167-178.

Zhang W, Peng G, Li C, et al. A new deep learning model for fault diagnosis with good anti-noise and domain adaptation ability on raw vibration signals[J]. Sensors, 2017, 17(2): 425.

Tong Z, Li W, Zhang B, et al. Bearing fault diagnosis based on domain adaptation using transferable features under different working conditions[J]. Shock and Vibration, 2018, 2018.

Pan S J, Tsang I W, Kwok J T, et al. Domain adaptation via transfer component analysis[J]. IEEE transactions on neural networks, 2010, 22(2): 199-210.

Zhang B, Li W, Li X L, et al. Intelligent fault diagnosis under varying working conditions based on domain adaptive convolutional neural networks[J]. Ieee Access, 2018, 6: 66367-66384.

Che C, Wang H, Fu Q, et al. Deep transfer learning for rolling bearing fault diagnosis under variable operating conditions[J]. Advances in Mechanical Engineering, 2019, 11(12): 1687814019897212.

Li X, Zhang W, Ding Q, et al. Multi-layer domain adaptation method for rolling bearing fault diagnosis[J]. Signal Processing, 2019, 157: 180-197.

Wang Q, Michau G, Fink O. Domain adaptive transfer learning for fault diagnosis[C]//2019 Prognostics and System Health Management Conference (PHM-Paris). IEEE, 2019: 279-285.

Guo L, Lei Y, Xing S, et al. Deep convolutional transfer learning network: A new method for intelligent fault diagnosis of machines with unlabeled data[J]. IEEE.

李彦冬, 郝宗波, 雷航. 卷积神经网络研究综述[J]. 计算机应用, 2016, 36(9):2508-2515

LeCun Y, Kavukcuoglu K, Farabet C. Convolutional networks and applications in vision[C]//Proceedings of 2010 IEEE international symposium on circuits and systems. IEEE, 2010: 253-256.

Nair V, Hinton G E. Rectified linear units improve restricted boltzmann machines[C]//Proceedings of the 27th international conference on machine learning (ICML-10). 2010: 807-814.

李理,应三丛.基于FPGA的卷积神经网络Softmax层实现[J].现代计算机(专业版),2017(26):21-24.

Ioffe S, Szegedy C. Batch normalization: Accelerating deep network training by reducing internal covariate shift[C]//International conference on machine learning. pmlr, 2015: 448-456.

Srivastava N, Hinton G, Krizhevsky A, et al. Dropout: a simple way to prevent neural networks from overfitting[J]. The journal of machine learning research, 2014, 15(1): 1929-1958.

Reddi S J, Kale S, Kumar S. On the convergence of adam and beyond[J]. arXiv preprint arXiv:1904.09237, 2019.

He K, Zhang X, Ren S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2016: 770-778.

Simonyan K, Zisserman A. Very deep convolutional networks for large-scale image recognition[J]. arXiv preprint arXiv:1409.1556, 2014.

Chorowski J K, Bahdanau D, Serdyuk D, et al. Attention-based models for speech recognition[J]. Advances in neural information processing systems, 2015, 28.

Vaswani A, Shazeer N, Parmar N, et al. Attention is all you need[J]. Advances in neural information processing systems, 2017, 30.

Dosovitskiy A, Beyer L, Kolesnikov A, et al. An image is worth 16x16 words: Transformers for image recognition at scale[J]. arXiv preprint arXiv:2010.11929, 2020.

Van der Maaten L, Hinton G. Visualizing data using t-SNE[J]. Journal of machine learning research, 2008, 9(11)