Endüstriyel IoT'de Tahmini Bakımın Geliştirilmesi: Basınç Anahtarlarında Gerçek Zamanlı Anomali Tespiti için Bir Makine Öğrenmesi Yaklaşımı

Yıl 2025, Cilt: 6 Sayı: 1, 1 - 9

Buket Soyhan , Zuhal Can

https://doi.org/10.53608/estudambilisim.1574345

Öz

Bu çalışma, endüstriyel bir ortamda basınç anahtarlarının davranışını izlemek ve tahmin etmek için makine öğrenimi tekniklerinin Nesnelerin İnterneti (IoT) verileriyle gerçek zamanlı entegrasyonunu araştırmıştır. Endüstriyel IoT sistemlerinde öngörücü bakım, operasyonel verimliliği artırmak ve beklenmeyen arızaları en aza indirmek için kritik öneme sahiptir. Endüstriyel süreçlerin artan karmaşıklığı, makine öğrenimi algoritmalarını ve IoT verilerini entegre etmeyi proaktif bakım için güçlü bir çözüm haline getirmiştir. Bu bağlamda çalışma, gerçek zamanlı sensör verilerini analiz ederek ve Rastgele Orman, Yalıtım Ormanı ve Yerel Aykırı Değer Faktörü algoritmalarını uygulayarak Basınç anahtarlarında anormallik tespiti ve arıza tahmini yapmayı amaçlamıştır. Modellerin performansı MetroPT-3 Tren Veri Seti üzerinde değerlendirilmiştir. Modellerin etkinliği, doğruluk, hassasiyet, geri çağırma ve F1 puanı gibi performans ölçütleriyle değerlendirilmiştir. Rastgele Orman Sınıflandırıcısı, %99,92'lik bir doğruluk oranıyla anormallik tespitinde en yüksek performansı göstermiştir. Bulgular, makine öğreniminin ve Nesnelerin İnterneti’nin öngörücü bakımı geliştirmede, sistem güvenilirliğini iyileştirmede ve daha geniş endüstriyel IoT alanına katkıda bulunmada önemli potansiyelini vurgulamıştır.

Anahtar Kelimeler

Öngörücü Bakım, Makine Öğrenimi, Anomali Algılama, Gerçek Zamanlı Sensör Verileri, Basınç Anahtarı İzleme

Enhancing Predictive Maintenance in Industrial IoT: A Machine Learning Approach for Real-time Anomaly Detection in Pressure Switches

Öz

This study investigates the real-time integration of machine learning techniques with Internet of Things (IoT) data to monitor and predict the behavior of pressure switches in an industrial environment. Predictive maintenance in industrial IoT systems is critical to increase operational efficiency and minimize unexpected failures. The increasing complexity of industrial processes has made integrating machine learning algorithms and IoT data a powerful solution for proactive maintenance. In this context, the study aims to perform anomaly detection and failure prediction in pressure switches by analyzing real-time sensor data and applying Random Forest, Isolation Forest, and Local Outlier Factor algorithms. The performance of the models is evaluated using the MetroPT-3 Train Dataset. Performance metrics like accuracy, precision, recall, and F1 score assess the models' effectiveness. The Random Forest Classifier showed the highest performance in anomaly detection with an accuracy rate of 99.92%. The findings emphasize the significant potential of machine learning and the Internet of Things in enhancing predictive maintenance, improving system reliability, and contributing to the broader field of industrial IoT.

Anahtar Kelimeler

Predictive Maintenance, Machine Learning, Anomaly Detection, Real-time Sensor Data

Etik Beyan

Ethical guidelines were followed in this study.

Destekleyen Kurum

The authors declare that they have no conflict of interest in this research.

Kaynakça

• Davari, N., Veloso, B., Ribeiro, R., Gama, J. 2023. MetroPT-3 Dataset. UCI Machine Learning Repository. Available: https://archive.ics.uci.edu/dataset/791/metropt+3+dataset
• Righetti, F., Vallati, C., Anastasi, G., Masetti, G., Di Giandomenico, F. 2020. Failure management strategies for IoT-based railways systems. 2020 IEEE International Conference on Smart Computing (SMARTCOMP), 386–391.
• Pradyumna, E. S. S., Sraavya, B., Chaithanya, J. K., Sreedhar, S. C., et al. 2021. Design and development of a prototype model for monitoring of automatic platform with train arrival using IoT. 2021 6th International Conference on Communication and Electronics Systems (ICCES), 707–712.
• Chatterjee, A., Ahmed, B. S. 2022. IoT anomaly detection methods and applications: A survey. Internet of Things, 19, 100568. DOI:10.1016/j.iot.2022.100568
• Silva, M. E. P., Veloso, B., Gama, J. 2023. Predictive maintenance, adversarial autoencoders and explainability. Joint European Conference on Machine Learning and Knowledge Discovery in Databases, 260–275.
• Cook, A. A., Mısırlı, G., Fan, Z. 2019. Anomaly detection for IoT time-series data: A survey. IEEE Internet of Things Journal, 7(7), 6481–6494. DOI: 10.1109/JIOT.2019.2958185
• Zhang, Y., Hao, Q., Cai, G., Lv, J., Yang, C. 2020. Crack damage identification and localisation on metro train bogie frame in IoT using guided waves. IET Intelligent Transport Systems, 14(11), 1403–1409. DOI:10.1049/iet-its.2020.0014
• Benazer, S. S., Dawood, M. S., Ramanathan, S. K., Saranya, G. 2021. Efficient model for IoT-based railway crack detection system. Materials Today: Proceedings, 45, 2789–2792. DOI:10.1016/j.matpr.2020.11.743
• Fraga-Lamas, P., Fernández-Caramés, T. M., Castedo, L. 2017. Towards the Internet of smart trains: A review on industrial IoT-connected railways. Sensors, 17(6), 1457. DOI:10.3390/s17061457
• Padhi, S., Subhedar, M., Behra, S., Patil, T. 2023. IoT-based condition monitoring for railway track fault detection in smart cities. IETE Journal of Research, 69(9), 5794–5803. DOI:10.1080/03772063.2022.2150701
• Karaduman, G., Karakose, M., Akin, E. 2018. Condition monitoring platform in railways based on IoT. 2018 International Conference on Artificial Intelligence and Data Processing (IDAP), 1–4.
• Siddiqui, H. U. R., Saleem, A. A., Raza, M. A., Zafar, K., Munir, K., Dudley, S. 2022. IoT-based railway track faults detection and localization using acoustic analysis. IEEE Access, 10, 106520–106533. DOI:10.1109/ACCESS.2022.3210326
• Zhao, Y., Yu, X., Chen, M., Zhang, M., Chen, Y., Niu, X., Sha, X., Zhan, Z., Li, W. J. 2020. Continuous monitoring of train parameters using IoT sensor and edge computing. IEEE Sensors Journal, 21(14), 15458–15468. DOI:10.1109/JSEN.2020.3026643
• Wang, H., Shi, T., Jiang, H. 2017. Railway freight safety comprehensive detection and tracking method based on IoT. 2017 5th International Conference on Frontiers of Manufacturing Science and Measuring Technology (FMSMT 2017), 494–501.
• Meng, Q., Lu, P., Zhu, S. 2023. A smartphone-enabled IoT system for vibration and noise monitoring of rail transit. IEEE Internet of Things Journal, 10(10), 8907–8917. DOI:10.1109/JIOT.2022.3233051
• Goodman, D. L., Hofmeister, J., Wagoner, R. 2015. Advanced diagnostics and anomaly detection for railroad safety applications: Using a wireless, IoT-enabled measurement system. 2015 IEEE AUTOTESTCON, 273–279. DOI:10.1109/AUTEST.2015.7356502
• Kebbeh, P. S., Seye, M. R., Ngom, B., Gueye, B., Diallo, M. 2018. Railmon: Distance, temperature and location railway monitoring using IoT technologies. Innovations and Interdisciplinary Solutions for Underserved Areas: Second International Conference, InterSol 2018, 212–223.
• Bilakanti, H., Pasam, S., Palakollu, V., Utukuru, S. 2024. Anomaly detection in IoT environment using machine learning. Security and Privacy, 7(3), e366. DOI:10.1002/spy2.366
• Singh, A., Aujla, G. S., Garg, S., Kaddoum, G., Singh, G. 2019. Deep-learning-based SDN model for Internet of Things: An incremental tensor train approach. IEEE Internet of Things Journal, 7(7), 6302–6311. DOI:10.1109/JIOT.2019.2953537
• Veloso, B., Ribeiro, R. P., Gama, J., Pereira, P. M. 2022. The MetroPT dataset for predictive maintenance. Scientific Data, 9(1), 764.
• Saki, M., Abolhasan, M., Lipman, J., Jamalipour, A. 2020. A comprehensive access point placement for IoT data transmission through train-wayside communications in multi-environment based rail networks. IEEE Transactions on Vehicular Technology, 69(10), 11937–11949. DOI:10.1109/TVT.2020.3006321
• Chellaswamy, C., Geetha, T. S., Vanathi, A., Venkatachalam, K. 2020. An IoT-based rail track condition monitoring and derailment prevention system. International Journal of RF Technologies, 11(2), 81–107. DOI:10.3233/RFT-190210
• Oransa, O., Abdel-Azim, M. 2015. ‘Railway as a Thing’: New railway control system in Egypt using IoT. 2015 Science and Information Conference (SAI), 124–133.
• Veloso, B., Gama, J., Ribeiro, R. P., Pereira, P. M. 2022. A benchmark dataset for predictive maintenance. arXiv Preprint, arXiv:2207.05466.
• Lakshmanna, K., Kaluri, R., Gundluru, N., Alzamil, Z. S., Rajput, D. S., Khan, A. A., Haq, M. A., Alhussen, A. 2022. A review on deep learning techniques for IoT data. Electronics, 11(10), 1604. DOI:10.3390/electronics11101604
• Sudharsan, B., Breslin, J. G., Ali, M. I. 2020. Edge2train: A framework to train machine learning models (SVMs) on resource-constrained IoT edge devices. Proceedings of the 10th International Conference on the Internet of Things, 1–8.
• Neamah, O. N., Bayır, R. 2024. Revolutionizing fault prediction in MetroPT datasets: Enhanced diagnosis and efficient failure prediction through innovative data refinement. 2024 Third International Conference on Power, Control and Computing Technologies (ICPC2T), 310–315.
• Barpute, J. V., Suryawanshi, S., Kshirsagar, V., Bhosale, D., Patil, P., Patil, A. 2024. Predictive maintenance of a metro’s air compressor. 2024 5th International Conference on Electronics and Sustainable Communication Systems (ICESC), 247–252.
• Nair, V., Premalatha, M., Braveen, M., et al. 2024. Enhancing metro rail efficiency: A predictive maintenance approach leveraging machine learning and deep learning technologies. DOI:10.21203/rs.3.rs-4319916/v1
• He, H., Ma, Y. 2013. Imbalanced learning: Foundations, algorithms, and applications. John Wiley & Sons.
• Fernández, A., García, S., Galar, M., Prati, R. C., Krawczyk, B., Herrera, F. 2018. Learning from imbalanced data sets. Springer.