Accuracy assessment of the One-Class SVM technique in identifying abnormalities in the vibration monitoring data of a gas turbine

Subject Areas : Vibration Transmision

Al-Tekreeti Watban Khalid Fahmi ¹ , Kazem Reza Kashyzadeh ^{2 *} , Siamak Ghorbani ³

1 - Department of Mechanical Engineering Technologies, Academy of Engineering, RUDN University, Moscow, Russia
2 - RUDN University
3 - Department of Mechanical Engineering Technologies, Academy of Engineering, RUDN University, Moscow, Russia

Received: 2024-03-21 Accepted : 2025-09-29 Published : 2026-01-09

Keywords: Power plant, Gas turbine, Fault detection, Vibration monitoring, Anomaly prediction.,

Abstract :

The existence of a strong and regular maintenance program in large industries such as power plants plays a vital role in preventing significant and irreparable damages to the system. In this article, the authors tried to evaluate the accuracy of one of the well-known methods, i.e., One-Class Support Vector Machine (SVM) technique, in the field of identifying anomalies related to the vibration data monitored in the power plant. For this purpose, a case study was conducted on the gas turbine of Kirkuk power plant located in Iraq. In this way, vibration monitoring was done by employing CA 202 piezoelectric accelerometer. After the analysis, the results indicate that the accuracy of this technique in detecting turbine vibration as a sign of a problem in the system is equal to 12.64%. These results emphasize the need for further research and modification of the proposed method because using the existing version of this technique without any development is ineffective to deal with the critical challenges in such industries, and in addition to not helping to reduce costs, it also leads to distrust of the system in other fault diagnosis methods.

References:

[1] A.T.W.K. Fahmi, K.R. Kashyzadeh, S. Ghorbani, A comprehensive review on mechanical failures cause vibration in the gas turbine of combined cycle power plants, Engineering Failure Analysis, Vo. 134, 2022. https://doi.org/10.1016/j.engfailanal.2022.106094.
[2] A.T.W.K. Fahmi, K.R. Kashyzadeh, S. Ghorbani, Industrial vibration detection techniques for enhanced monitoring and maintenance of combined cycle power plants, in International Conference on Engineering Systems, Moscow, Russian Federation, 5-7 April 2023.
[3] C. Scheffer, P. Girdhar, Practical machinery vibration analysis and predictive maintenance, 1st Edition, Newnes: Elsevier, 2004.
[4] S. Karmakar, S. Chattopadhyay, M. Mitra, S. Sengupta, Induction motor fault diagnosis: General discussion and research scope, in Karmakar S, Chattopadhyay S, Mitra M and Sengupta S. (eds) Induction Motor Fault Diagnosis, Singapore: Springer, 2016. https://doi.org/10.1007/978-981-10-0624-1_9.
[5] S. Zare, Fault detection and diagnosis of electric drives using intelligent machine learning approaches, PhD Thesis, University of Windsor, Canada, 2018.
[6] D. Goyal, BS. Pabla, The vibration monitoring methods and signal processing techniques for structural health monitoring: a review, Arch Comput Methods Eng., Vol. 23, No. 4, pp. 585–594, 2016. https://doi.org/10.1007/s11831-015-9145-0.
[7] S.J. Lacey, The Role of Vibration Monitoring in Predictive Maintenance Part 1-Principles and Practice, Maintenance & Asset Management. Maintenance & Engineering Magazine, Vol. 24, No. 1, pp. 42-51, 2011‏.
[8] RK. Mobley, 50 - Predictive Maintenance, Mobley RK (eds) Plant Engineer’s Handbook, 1st Edition, Butterworth: Elsevier, 2001.
[9] E. Losi, M. Venturini, L. Manservigi, GF. Ceschini, G. Bechini, Anomaly detection in gas turbine time series using bayesian hierarchical models, J Eng Gas Turbines Power, Vol. 141, No. 11, 2019. https://doi.org/10.1115/1.4044781.
[10] GP. Zhang, Time series forecasting using a hybrid ARIMA and neural network model, Neurocomputing, Vol. 50, pp. 159–175, 2003. https://doi.org/10.1016/S0925-2312(01)00702-0.
[11] K. Koc, Ö. Ekmekcioğlu, AP. Gurgun, Accident prediction in construction using hybrid wavelet-machine learning, Automation in Construction, Vol. 133, 2022. https://doi.org/10.1016/j.autcon.2021.103987.
[12] D. Solyali, A comparative analysis of machine learning approaches for short-/long-term electricity load forecasting in Cyprus, Sustainability, Vol. 12, No. 9, 2020. https://doi.org/10.3390/su12093612.
[13] G. Mahalakshmi, S. Sridevi, S. Rajaram, A survey on forecasting of time series data. in International Conference on Computing Technologies and Intelligent Data Engineering (ICCTIDE'16), Kovilpatti, India, 7-9 January 2016. https://doi.org/10.1109/ICCTIDE.2016.7725358.
[14] M. Huang, Z. Liu, Research on mechanical fault prediction method based on multi-feature fusion of vibration sensing data, Sensors, Vol. 20, No. 1, 2019. https://doi.org/10.3390/s20010006.
[15] S. Gawde, S. Patil, S. Kumar, P. Kamat, K. Kotecha, A. Abraham, Multi-fault diagnosis of Industrial Rotating Machines using Data-driven approach: A review of two decades of research, Engineering Applications of Artificial Intelligence, Vol. 123, 2023. https://doi.org/10.1016/j.engappai.2023.106139.
[16] Y.E. Lee, B.K. Kim, J.H. Bae, K.C. Kim, Misalignment detection of a rotating machine shaft using a support vector machine learning algorithm, International Journal of Precision Engineering and Manufacturing, Vol. 22, pp. 409-416, 2021. https://doi.org/10.1007/s12541-020-00462-1.
[17] E. Maleki, O. Unal, S.S. Sahebari, K.R. Kashyzadeh, A Novel Approach for Analyzing the Effects of Almen Intensity on the Residual Stress and Hardness of Shot-Peened (TiB+ TiC)/Ti–6Al–4V Composite: Deep Learning, Materials, Vol. 16, 2023. ‏ https://doi.org/10.3390/ma16134693.
[18] K.R. Kashyzadeh, S. Ghorbani, New neural network-based algorithm for predicting fatigue life of aluminum alloys in terms of machining parameters, Engineering Failure Analysis, Vol. 146, 2023.‏ https://doi.org/10.1016/j.engfailanal.2023.107128.
[19] E. Maleki, O. Unal, S.M.S. Sahebari, K.R. Kashyzadeh, N. Amiri, Enhancing Friction Stir Welding in Fishing Boat Construction through Deep Learning-Based Optimization, Sustainable Marine Structures, Vol. 5, No. 2, pp. 1-14, 2023. https://doi.org/10.36956/sms.v5i2.875.
[20] Q. Wang, H. Qi, F. Liu, Time series prediction of E-nose sensor drift based on deep recurrent neural network, in Chinese Control Conference (CCC), Guangzhou, China, 27-30 July 2019. https://doi.org/10.23919/ChiCC.2019.8866168.
[21] A.T.W.K. Fahmi, K.R. Kashyzadeh, S. Ghorbani, Fault Detection in the Gas Turbine of the Kirkuk Power Plant: An Anomaly Detection Approach Using DLSTM-Autoencoder, Engineering Failure Analysis, Vol. 160, 2024. https://doi.org/10.1016/j.engfailanal.2024.108213.
[22] M. Amer, M. Goldstein, S. Abdennadher, Enhancing one-class support vector machines for unsupervised anomaly detection, in Proceedings of the ACM SIGKDD workshop on outlier detection and description, 2013. https://doi.org/10.1145/2500853.2500857.