Health Monitoring and Predictive Maintenance of Platform Screen Door Systems Using Artificial Intelligence

Abstract

The Platform Screen Door System (PSD) is a sliding door system used as a barrier between passengers and the vehicle/rail in modern metro and Rapid Bus Transit (RBT) stations. The PSD system not only serves as a barrier between the platform and tracks but also provides safe boarding and alighting opportunities for passengers, making it a critical component of modern metro stations. Consequently, PSD systems have rapidly gained popularity and are widely used. In recent years, numerous research studies have been conducted on PSD systems, covering a broad range of topics such as station environment conditions, energy consumption, passenger waiting times, emergency evacuation procedures, safety-SIL procedures, and control and monitoring approaches for PSD systems. Continued research and development of PSD systems is necessary due to their critical importance for passenger safety and their indispensable role in modern metro stations. Machine learning algorithms have played a significant role in fault diagnosis, and these algorithms can be used to improve the reliability of PSD systems. The results of studies conducted using these fault diagnosis methods could help in real-time detection and rectification of errors by monitoring system performance. Artificial intelligence-based predictive maintenance approaches are important, particularly in the railway sector, for enhancing both passenger safety and operational performance. This study focuses on the application of artificial intelligence models, such as Support Vector Machine (SVM), K-Nearest Neighbors (KNN), and Logistic Regression (LR), for the diagnosis of mechanical failures in full-scale PSD systems. The study utilized data such as current and voltage information, vibration, sound, door position, and door speed provided by the PSD system. The features of these data were extracted and used in machine learning algorithms to diagnose faults that could occur in the system.

Keywords

• ITS
• Platform Screen Door
• Predictive Maintenance
• Artificial Intelligence
• Machine Learning
• Condition Monitoring

Project Number

9210043

Yapay Zekâ Kullanımıyla Peron Ayırıcı Kapı Sisteminin Sağlığını İzleme ve Kestirimci Bakım

Abstract

Peron Ayırıcı Kapı Sistemi (PAKS), modern metro ve Hızlı Otobüs Transit (RBT) istasyonlarında yolcu ve araç/ray arasında bir bariyer olarak kullanılan kayar kapı sistemleridir. PAKS sistemi, sadece platform ve raylar arasında bir bariyer olmakla kalmaz, aynı zamanda araçlara emniyetli iniş ve biniş imkânı da sağlar. Bu nedenle, günümüzde PAKS sistemi metro istasyonlarında hızla yaygınlaşmakta ve kullanılmaktadır. Son yıllarda, PAKS sistemi ile ilgili birçok çalışma yapılmıştır. Bu çalışmalar, istasyon çevresi koşullarından, enerji tüketimine, yolcu bekleme sürelerine, acil tahliye prosedürlerine, emniyet-SIL prosedürlerine ve PAKS sisteminin kontrol ve izleme yaklaşımlarına kadar geniş bir yelpazeyi kapsamaktadır. PAKS sistemi, yolcu emniyeti için kritik bir önem taşımakta ve modern metro istasyonlarının vazgeçilmez bir özelliği haline gelmiştir. Bu nedenle, PAKS sistemi üzerine yapılan araştırmaların devam etmesi ve sistemin sürekli olarak geliştirilmesi gereklidir. Makine öğrenimi algoritmaları, hata teşhisinde önemli bir katkı sağlamakta ve bu algoritmalar sayesinde sistemin sürekli olarak geliştirilmesi hedeflenmektedir. Hata teşhisi yöntemleri kullanılarak gerçekleştirilen çalışmaların sonuçları, sistem performansını gerçek zamanlı olarak izleyerek hataların tespit edilmesine ve giderilmesine yardımcı olmaktadır. Yapay zekâ tabanlı öngörülü bakım yaklaşımı, özellikle demiryolu sektöründe hem yolcu emniyetini hem de işletme performansını artırmak için önemlidir. Bu çalışma, tam boy PAKS sistemlerinde makine öğrenmesi tabanlı sınıflandırma modellerinin kullanımı (SVM, KNN ve LR) ile mekanik arızaların teşhisini içermektedir. Çalışmada, PAKS sistemi tarafından sağlanan akım, gerilim, titreşim, ses, kapı pozisyonu ve kapı hızı gibi veriler kullanılmıştır. Bu verilerin istatistiksel öznitelikleri çıkarılmış ve bu öznitelikler makine öğrenimi algoritmalarında kullanılarak sistemdeki arızaların tespiti yapılmıştır.

Keywords

• AUS
• Peron Ayırıcı Kapı Sistemleri
• Kestirimci Bakım
• Yapay Zekâ
• Makine Öğrenmesi
• Durum İzleme

Supporting Institution

TÜBİTAK

Project Number

9210043

Thanks

Bu çalışma, ECOMAI PENTA-EURIPIDES (Hibe No. 2021028) çatı projesi kapsamında TÜBİTAK (Hibe No. 9210043) tarafından desteklenmiştir.

References

1. Koç, İ., Mermer, Ö., Kırımça, N., Çakır, F.H., ve Karaköse, M. (2022). Modeling and Simulation of Platform Screen Door (PSD) System using MATLAB-Simulink. International Conference on Data Analytics for Business and Industry (ICDABI), Sakhir – Kingdom of Bahrain, 629-633.
2. Li, X., ve Wang, Y., (2018). Simulation study on air leakage of platform screen doors in subway stations. Sustainable Cities and Society, c. 43, 350-356.
3. Zhou, C., Su, Z., ve Zhou, J. (2010). Design and Implementation of the Platform Screen Doors System for BRT, 2540-2552. doi: 10.1061/41127(382)271.
4. Abdurrahman, U.T., Jack, A., ve Schmid, F. (2018). Effects of Platform Screen Doors on the Overall Railway System. 8th International Conference on Railway Engineering, London, UK. doi: 10.1049/cp.2018.0053.
5. Roh, J. S., Ryou, H.S., ve Yoon, S.W. (2010). The effect of PSD on life safety in subway station fire. J Mech Sci Technol, 24(4), 937-942. doi: 10.1007/s12206-010-0217-7.
6. Qu, L., ve Chow, W. K., (2012). Platform siren doors on emergency evacuation in underground railway stations. Tunnelling and Underground Space Technology, 30(1), 1-9. doi: 10.1016/j.tust.2011.09.003.
7. Lindfeldt, O., (2017). The impact of platform screen doors on rail capacity, Int. J. TDI, 1(3), 601-610. doi: 10.2495/TDI-V1-N3-601-610
8. Su, Z., ve Li, X., (2022). Energy benchmarking analysis of subway station with platform screen door system in China. Tunnelling and Underground Space Technology, 128, 104655. doi: 10.1016/j.tust.2022.104655.

9. Gabay, D., (2004) Compared fire safety features for metro tunnels, Safe & Reliable Tunnels. Innovative European Achievements First International Symposium, 4-6 February, Prague.
10. Aarnio, P., Yli-Tuomi, T., Kousa, A., Mäkelä, T., Hirsikko, A., Hämeri, K., Räisänen, M., Hillamo, R., Koskentalo, T., Jantunen, M., (2005). The concentrations and composition of and exposure to fine particle in the Helsinki subway system. Atmos. Environ. 39(28), 5059–5066.
11. Ampofo, F., Maidment, G., Missenden, J., 2004. Underground railway environment in the UK Part 1: Review of thermal comfort. Appl. Therm. Eng. 24 (5), 611–631.
12. He, S., Jin, L., Le, T., Zhang, C., Liu, X., ve Ming, X., (2018). Commuter health risk and the protective effect of three typical metro environmental control systems in Beijing, China. Transportation Research Part D: Transport and Environment, c. 62, 633-645.
13. Min, L., Zhaoyong, C., ve Jin, Z., (2012). Study on PSD system control strategy for safety. 3rd International Conference on System Science, Engineering Design and Manufacturing Informatization, Chengdu, 154-159. doi: 10.1109/ICSSEM.2012.6340789.
14. Koç, İ., Mermer, Ö., Kırımça, N., ve Karaköse, M. (2023). Raylı Sistemlerde Peron Ayırıcı Kapı Sistemi İçin Yapay Sinir Ağı Tabanlı Hata Teşhis Yaklaşımı. Emo Bilimsel Dergi, 13(1), 13-22.
15. Li, C., Luo, S., Cole, C., ve Spiryagin, M., (2017). An overview: modern techniques for railway vehicle on-board health monitoring systems, Vehicle System Dynamics, c. 55(7), 1045-1070.
16. Gonzalez-Jimenez, D., Del-Olmo J., Poza, J., Garramiola, F., ve Madina P., (2021). Data-Driven Fault Diagnosis for Electric Drives: A Review, Sensors, c. 21(12), s. 4024.
17. Sun, X., Ling, K. V., Sin, K. K., ve Tay, L. (2018). Intelligent Fault Detection and Diagnosis of Air Leakage on Train Door. International Conference on Intelligent Rail Transportation (ICIRT), Singapore, 1(4).
18. Başaran, M., Fidan, M., (2020). Gearbox Fault Classification by Using Frequency Based Feature Extraction. Eskişehir Technical University Journal of Science and Technology, 21, 101-107.
19. Ham, S., Han, S.Y., Kim, S., Park, H. J., Park, K. J., ve Choi J. H. (2019). A Comparative Study of Fault Diagnosis for Train Door System. Traditional versus Deep Learning Approaches, Sensors, c. 19(23) s. 5160.
20. Deng, L., ve Yu, D. (2014). Deep learning: methods and applications. Foundations and trends® in Signal Processing, 7(3–4), 197-387.
21. Mimaz M. R., Yıldız, K. (2019). İndiksiyon Motorun Mekanik Arıza Teşhisinde Makine Öğrenme Yöntemlerinin Kullanılması. Avrupa Bilim ve Teknoloji Dergisi, Sayı 16, S. 881-904.
22. Zagajewski, B., Kluczek, M., Raczko, E., Njegovec, A., Dabija, A., & Kycko, M. (2021). Comparison of random forest, support vector machines, and neural networks for post-disaster forest species mapping of the krkonoše/karkonosze transboundary biosphere reserve. Remote Sensing, 13(13), 2581.
23. Zoppis, I., Mauri, G., & Dondi, R. (2019). Kernel methods: Support vector machines. In Encyclopedia of Bioinformatics and Computational Biology. Volume 1 (pp. 503-510). Elsevier.
24. Ben-Hur, A., Horn, D., Siegelmann, H. T., & Vapnik, V. (2001). Support vector clustering. Journal of Machine Learning Research, 2(Dec), 125-137.
25. Hsieh, C. J., Chang, K. W., Lin, C. J., Keerthi, S. S., & Sundararajan, S. (2008). A dual coordinate descent method for large-scale linear SVM. Proceedings of the 25th International Conference on Machine Learning (pp. 408-415)
26. Khan, M. M. R., Arif, R. B., Siddique, M. A. B., & Oishe, M. R. (2018). Study and observation of the variation of accuracies of KNN, SVM, LMNN, ENN algorithms on eleven different datasets from UCI machine learning repository. 4th International Conference on Electrical Engineering and Information & Communication Technology (iCEEiCT) (pp. 124-129). IEEE.
27. Joshuva, A., Sugumaran, V., & Amarnath, M. (2015). Selecting kernel function of support vector machine for fault diagnosis of roller bearings using sound signals through histogram features. International Journal of Applied Engineering Research, 10(68), 482-487.
28. Pandya, D., Upadhyay, S. H., & Harsha, S. P. (2014). Fault diagnosis of rolling element bearing by using multinomial logistic regression and wavelet packet transform. Soft Computing, 18, 255-266.
29. Shuai, L., Limin, J., Yong, Q., Bo, Y., & Yanhui, W. (2014). Research on urban rail train passenger door system fault diagnosis using PCA and rough set. The Open Mechanical Engineering Journal, 8(1).
30. Sun, L., Zhang, J., Ding, W., & Xu, J. (2022). Feature reduction for imbalanced data classification using similarity-based feature clustering with adaptive weighted K-nearest neighbors. Information Sciences, 593, 591-613.