Fault detection for rotating machinery based on vibration data using machine learning

• Lucas de T. Barreto Federal University of Santa Catarina
• Rodrigo K. Rosa Federal University of Santa Catarina
• Danilo Silva Federal University of Santa Catarina
• Danilo Braga Dynamox

DOI: https://doi.org/10.5753/eniac.2023.233935

Abstract

This paper addresses the detection of mechanical faults in rotating machinery using machine learning techniques. Vibration signals were collected from machines in operation in the industry, and features of these signals were extracted, ranging from harmonics of a motor's rotation speed to specialized features typically considered by vibration analysts. After data cleaning, preprocessing, and the construction of the training pipeline and hyperparameter optimization, machine learning models such as logistic regression, support vector machines, random forests, neural networks, and gradient boosting (XGBoost) were explored. The results showed that the XGBoost model performed the best, achieving an ROC AUC metric of 91%.

Keywords: Fault detection, vibration analysis, rotating machine, machine learning

References

Ahmed, H. and Nandi, A. K. (2020). Condition monitoring with vibration signals: Compressive sampling and learning algorithms for rotating machines. John Wiley & Sons.

Alharbi, F., Luo, S., Zhang, H., Shaukat, K., Yang, G., Wheeler, C. A., and Chen, Z. (2023). A brief review of acoustic and vibration signal-based fault detection for belt conveyor idlers using machine learning models. Sensors, 23(4):1902.

Bishop, C. M. (2006). Pattern Recognition and Machine Learning (Information Science and Statistics). Springer-Verlag, Berlin, Heidelberg.

Boashash, B. (2015). Time-frequency signal analysis and processing: a comprehensive reference. Academic press.

Cervantes, J., Garcia-Lamont, F., Rodríguez-Mazahua, L., and Lopez, A. (2020). A comprehensive survey on support vector machine classification: Applications, challenges and trends. Neurocomputing, 408:189–215.

Chen, T. and Guestrin, C. (2016). XGBoost. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. ACM.

Gao, Z., Cecati, C., and Ding, S. X. (2015). A survey of fault diagnosis and fault-tolerant techniques—part i: Fault diagnosis with model-based and signal-based approaches. IEEE transactions on industrial electronics, 62(6):3757–3767.

Géron, A. (2019). Hands-on Machine Learning with Scikit-Learn, Keras, and Tensor-Flow: Concepts, Tools, and Techniques to Build Intelligent Systems. O’Reilly Media, Incorporated.

Janssens, O., Slavkovikj, V., Vervisch, B., Stockman, K., Loccufier, M., Verstockt, S., Van de Walle, R., and Van Hoecke, S. (2016). Convolutional neural network based fault detection for rotating machinery. Journal of Sound and Vibration, 377:331–345.

Laguna, C. and Lerch, A. (2016). An efficient algorithm for clipping detection and declipping audio. In Audio Engineering Society Convention 141. Audio Engineering Society.

Lei, Y., Yang, B., Jiang, X., Jia, F., Li, N., and Nandi, A. K. (2020). Applications of machine learning to machine fault diagnosis: A review and roadmap. Mechanical Systems and Signal Processing, 138:106587.

Randall, R. B. (2021). Vibration-based condition monitoring: industrial, automotive and aerospace applications. John Wiley & Sons.

Welch, P. D. (1967). The use of fast fourier transforms for the estimation of power spectra: A method based on time averaging over short modified periodograms. IEEE Transactions on Audio and Electroacoustics, 15:70–73.

Zimroz, R. and Król, R. (2009). Failure analysis of belt conveyor systems for condition monitoring purposes. Mining Science, 128(36):255.