Kısıtlı kestirimci bakım için makine öğrenmesi yöntemlerinin kıyası

Yıl 2025, , 183 - 191, 15.01.2025

Timur Ozkul , Ayça Topallı

https://doi.org/10.28948/ngumuh.1465282

Öz

Kestirimci bakım, duyaçların varlığı ve teçhizatların bağlanabilirliği ile son zamanlarda artan bir ilgi elde etmiştir. Yine de, özellikle eski cihazlardan geniş çapta veri elde etmek zor olabilir. Bu makale, verilerin endüstriyel bir düzenekten alınan alarm kayıtları ile sınırlı olduğu bir ortam için, geçmiş bilgileri kullanarak yakın gelecekteki bir durumu öngören akıllı bir yöntemi tanımlamaktadır. Makine öğrenmesi yöntemlerinin zaman dizisi verileri kullanarak sınıflandırma yapma işlerinde etkili olduğu kanıtlanmış olduğundan, sinir ağları, rassal orman ve aşırı eğim arttırma olarak seçilen üç yöntem, bir alarmın ve aynı makinenin kaydettiği diğer alarmların geçmiş oluşumlarından, o alarmı iki saat önceden tahmin etmek üzere eğitilmiştir. Bu üç yöntemin performansları kıyaslanmış ve hiper-parametre değerleri arasından en iyi yapılandırmayı bulmak hedeflenmiştir. Elde edilen sonuçlara göre, aşırı eğim arttırma, 500 ağaç sayısı, 128 azami derinlik ve son günden alarm oluşumları girdi penceresi ile 0.767 olan en yüksek F1 puanını vermektedir. Bu çalışma, makinelerin işlemesi ve bakımı hakkında potansiyel olarak önemli anlayışlar sağlayan ve dikkate değer masraf azaltma imkânları sunan alarm öngörüleri için en iyi makine öğrenmesi yöntemini belirlemeyi hedefleyen kıyaslamalı bir araştırmadan oluşmaktadır.

Anahtar Kelimeler

Makine Öğrenmesi, Sinir Ağları, Kestirimci Bakım, Rassal Orman, Aşırı Eğim Arttırma

Comparison of machine learning methods for limited predictive maintenance

Öz

Predictive maintenance has gained increasing attention recently with the availability of sensors and connectivity of equipment. Yet, it would be difficult to obtain a wide range of data, especially with legacy devices. This paper describes an intelligent method for predicting a near future condition using the past information for an environment in which data are limited to the alarm logs from industrial machinery. Since machine learning methods are proven to be efficient in classification tasks using time series data, three of them are selected to predict an alarm two hours in advance using the past occurrences. These methods are neural networks, random forests, and extreme gradient boosting. The performances of these three methods are compared, and it is aimed to find the optimal configuration among hyper-parameter values. According to the obtained results, extreme gradient boosting gives the highest F1-score of 0.767 with number of trees equal to 500, maximum depth of 128, and an input window of alarm occurrences from the last day. This work consists of a comparative study aiming to identify the best machine learning method for alarm predictions, which potentially provides important insights into the operation and maintenance of machinery, bringing the possibility of considerable cost reductions.

Anahtar Kelimeler

Machine Learning, Neural Networks, Predictive Maintenance, Random Forest, Extreme Gradient Boosting

Kaynakça

• H.M. Hashemian and W.C. Bean, State-of-the-Art Predictive Maintenance Techniques, IEEE Trans. Instrum. Meas., 60, 3480–3492, 2011. https://doi.org/ 10.1109/tim.2009.2036347.
• K.T.P. Nguyen and K. Medjaher, A new dynamic predictive maintenance framework using deep learning for failure prognostics, Reliability Engineering & System Safety, 188, 251–262, 2019. https://doi.org/ 10.1016/j.ress.2019.03.018.
• I.J. Akpan, E.A.P. Udoh and B. Adebisi, Small business awareness and adoption of state-of-the-art technologies in emerging and developing markets, and lessons from the COVID-19 pandemic, Journal of Small Business & Entrepreneurship, 34, 123–140, 2020. https://doi.org/10.1080/08276331.2020.182018 5.
• T.P. Carvalho, F.A.A.M.N. Soares, R. Vita, R. da P. Francisco, J.P. Basto and S.G.S. Alcalá, A systematic literature review of machine learning methods applied to predictive maintenance, Computers & Industrial Engineering, 137, 106024, 2019. https://doi.org/ 10.1016/j.cie.2019.106024.
• S. Ayvaz and K. Alpay, Predictive maintenance system for production lines in manufacturing: A machine learning approach using IoT data in real-time, Expert Systems with Applications, 173, 114598, 2021. https:// doi.org/10.1016/j.eswa.2021.114598.
• T. Wuest, D. Weimer, C. Irgens and K.-D. Thoben, Machine learning in manufacturing: advantages, challenges, and applications, Production & Manufacturing Research, 4, 23–45, 2016. https:// doi.org/10.1080/21693277.2016.1192517.
• W. Li, H. Li, S. Gu and T. Chen, Process fault diagnosis with model- and knowledge-based approaches: Advances and opportunities, Control Engineering Practice, 105, 104637, 2020. https://doi.org/10.1016 /j.conengprac.2020.104637.
• Y. He, C. Gu, Z. Chen and X. Han, Integrated predictive maintenance strategy for manufacturing systems by combining quality control and mission reliability analysis, International Journal of Production Research, 55, 5841–5862, 2017. https://doi.org/10. 1080/00207543.2017.1346843.
• W. Yu, T. Dillon, F. Mostafa, W. Rahayu and Y. Liu, A Global Manufacturing Big Data Ecosystem for Fault Detection in Predictive Maintenance, IEEE Trans. Ind. Inf., 16, 183–192, 2020. https://doi.org/10.1109/tii.201 9.2915846.
• O. Güler, Turbofan motorlarının kestirimci bakımında makine öğrenimi algoritmaları performanslarının karşılaştırılması, NOHU J. Eng. Sci., 13, 1, 99–106, 2024. https://doi.org/doi: 10.28948/ngumuh.1266541.
• G. Dorgo and J. Abonyi, Sequence Mining Based Alarm Suppression, IEEE Access, 6, 15365–15379, 2018. https://doi.org/10.1109/access.2018.2797247.
• E. Ruschel, E.A.P. Santos and E. de F.R. Loures, Industrial maintenance decision-making: A systematic literature review, Journal of Manufacturing Systems, 45, 180–194, 2017. https://doi.org/10.1016/j.jmsy.201 7.09.003.
• M. Baptista, S. Sankararaman, Ivo.P. de Medeiros, C. Nascimento Jr., H. Prendinger and E.M.P. Henriques, Forecasting fault events for predictive maintenance using data-driven techniques and ARMA modeling, Computers & Industrial Engineering, 115, 41–53, 2018. https://doi.org/10.1016/j.cie.2017.10.033.
• A. Bousdekis, K. Lepenioti, D. Apostolou and G. Mentzas, Decision Making in Predictive Maintenance: Literature Review and Research Agenda for Industry 4.0, IFAC-PapersOnLine, 52, 607–612, 2019. https:// doi.org/10.1016/j.ifacol.2019.11.226.
• J.-H. Shin, H.-B. Jun and J.-G. Kim, Dynamic control of intelligent parking guidance using neural network predictive control, Computers & Industrial Engineering, 120, 15–30. 2018 https://doi.org/10.10 16/j.cie.2018.04.023.
• S. Gomes Soares and R. Araújo, An on-line weighted ensemble of regressor models to handle concept drifts, Engineering Applications of Artificial Intelligence, 37, 392–406, 2015. https://doi.org/10.1016/j.engappai.201 4.10.003.
• D.D. Pezze, C. Masiero, D. Tosato, A. Beghi and G.A. Susto, FORMULA: A Deep Learning Approach for Rare Alarms Predictions in Industrial Equipment, IEEE Trans. Automat. Sci. Eng., 19, 1491–1502, 2022. https://doi.org/10.1109/tase.2021.3127995.
• D. Tosato, D. Dalle Pezze, C. Masiero, G.A. Susto and A. Beghi, Alarm logs of industrial packaging machines, 2020. https://doi.org/10.21227/NFV6-K750.
• N. Kolokas, T. Vafeiadis, D. Ioannidis and D. Tzovaras, Forecasting faults of industrial equipment using machine learning classifiers, 2018 Innovations in Intelligent Systems and Applications (INISTA), 2018. https://doi.org/10.1109/inista.2018.8466309.
• S. Biswal and G.R. Sabareesh, Design and development of a wind turbine test rig for condition monitoring studies, 2015 International Conference on Industrial Instrumentation and Control (ICIC), 2015. https://doi.org/10.1109/iic.2015.7150869.
• J. Zhu, C. Wang, C. Li, X. Gao and J. Zhao, Dynamic alarm prediction for critical alarms using a probabilistic model, Chinese Journal of Chemical Engineering, 24, 881–885, 2016. https://doi.org/10.1016/j.cjche.2016.0 4.017.
• R. Prytz, S. Nowaczyk, T. Rögnvaldsson and S. Byttner, Predicting the need for vehicle compressor repairs using maintenance records and logged vehicle data, Engineering Applications of Artificial Intelligence, 41, 139–150, 2015. https://doi.org/10.10 16/j.engappai.2015.02.009.
• K. Kulkarni, U. Devi, A. Sirighee, J. Hazra and P. Rao, Predictive Maintenance for Supermarket Refrigeration Systems Using Only Case Temperature Data, 2018 Annual American Control Conference (ACC), 2018. https://doi.org/10.23919/acc.2018.8431901.
• T. dos Santos, F.J.T.E. Ferreira, J.M. Pires and C. Damasio, Stator winding short-circuit fault diagnosis in induction motors using random forest, 2017 IEEE International Electric Machines and Drives Conference (IEMDC), 2017. https://doi.org/10.1109/iemdc.2017.8 002350.
• M. Paolanti, L. Romeo, A. Felicetti, A. Mancini, E. Frontoni and J. Loncarski, Machine Learning approach for Predictive Maintenance in Industry 4.0, 2018 14th IEEE/ASME International Conference on Mechatronic and Embedded Systems and Applications (MESA), 2018. https://doi.org/10.1109/mesa.2018.8449150.
• C.-J. Su and S.-F. Huang, Real-time big data analytics for hard disk drive predictive maintenance, Computers & Electrical Engineering, 71, 93–101, 2018. https://doi.org/10.1016/j.compeleceng.2018.07.025.
• B. Steurtewagen and D. Van den Poel, Adding interpretability to predictive maintenance by machine learning on sensor data, Computers & Chemical Engineering, 152, 107381, 2021. https://doi.org/10.10 16/j.compchemeng.2021.107381.
• L.S. Shapley, 17. A Value for n-Person Games, Contributions to the Theory of Games (AM-28), Volume II, 307–318, 1953. https://doi.org/10.1515/97 81400881970-018.
• L. Breiman, Machine Learning 45, 5–32, 2001. https://doi.org/10.1023/a:1010933404324.
• C. Strobl, J. Malley and G. Tutz, An introduction to recursive partitioning: Rationale, application, and characteristics of classification and regression trees, bagging, and random forests., Psychological Methods, 14, 323–348, 2009. https://doi.org/10.1037/a0016973.
• T. Chen and C. Guestrin, XGBoost, Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2016. https:// doi.org/10.1145/2939672.2939785.
• D. Nielsen, Tree Boosting With XGBoost - Why Does XGBoost Win ‘Every’ Machine Learning Competition? Master Thesis, Norwegian University of Science and Technology, Department of Mathematical Sciences, Norway, 2016.
• H. Belyadi and A. Haghighat, Supervised learning, in Machine Learning Guide for Oil and Gas Using Python, Gulf Professional Publishing, 2021, pp. 169-295. https://doi.org/10.1016/B978-0-12-821929-4.000 04-4.