Intellectual Technologies on Transport Intellectual Technologies on Transport Интеллектуальные технологии на транспорте 2413-2527 123840 10.20295/2413-2527-2026-246-82-90 МАТЕМАТИЧЕСКОЕ МОДЕЛИРОВАНИЕ И СИСТЕМНЫЙ АНАЛИЗ MATHEMATICAL MODELLING AND SYSTEM ANALYSIS МАТЕМАТИЧЕСКОЕ МОДЕЛИРОВАНИЕ И СИСТЕМНЫЙ АНАЛИЗ Contrastive Self‑Supervised Learning in Transportation Monitoring Systems: A Practical Approach to Early Anomaly Detection Контрастивное самообучение в системах мониторинга транспорта: практический подход к ранней Божко Леся Михайловна Bozhko Lesya Mikhailovna lemib@rambler.ru

доктор экономических наук;

doctor of economic sciences;

 Коньков Дмитрий Александрович Konkov Dmitry Alexandrovich dmitry.konkov@letka.space Петербургский государственный университет путей сообщения Императора Александра I Россия Emperor Alexander I St. Petersburg State Transport University Russian Federation Петербургский государственный университет путей сообщения Императора Александра I Россия Emperor Alexander I St. Petersburg State Transport University Russian Federation 24 06 2026 24 06 2026 2 82 90 20 05 2026 26 05 2026 https://itt-pgups.ru/en/nauka/article/123840/view

Для выявления дефектов и предупреждения аварий в транспортных системах используются датчики, применяющие модели на основе классических конструкций или пороговых правил, что чревато ложными тревогами или запаздыванием. Модели контрастивного самообучения позволяют снять указанные недостатки, но требуют доработки под прикладные задачи. Цель исследования: предложить и описать практический подход к использованию методов контрастивного самообучения для задач раннего обнаружения аномалий в системах мониторинга транспорта. Методы: применены методы контрастивного обучения с функцией потерь InfoNCE, легковесные сверточные энкодеры (1D-CNN/1D-ResNet), вариационные автокодировщики (VAE) для реконструкционного контроля, механизмы внимания для оценки вкладов сенсоров, адаптивная калибровка порогов на основе экспоненциального скользящего среднего (EMA), кластеризация представлений для множественных нормальных режимов и сценарии тестирования на полуреальных данных. Результаты: выявлены особенности мультисенсорных временных рядов, описаны ограничения полевых устройств и определены пути уменьшения ложных срабатываний при ограниченной разметке. Предложены алгоритмические блоки для практического внедрения: аугментации временных рядов, адаптивная калибровка порогов, attention-механизмы для объяснимости и валидация на полуреальных данных. Разработан гибридны критерий аномальности контрастивной и реконструкционной оценке аномальности. Практическая значимость: внедрение контрастивного блока с адаптивной калибровкой обеспечивает снижение времени обнаружения аномалий и сокращение частоты ложных срабатываний по сравнению с базовыми VAE и пороговыми системами. Обсуждение: рекомендуется интеграция контрастивных модулей и адаптивной калибровки в существующие системы мониторинга транспорта с использованием трехуровневой системы тревог и объяснений по каналам.

To detect defects and prevent accidents in transport systems, sensors are used that use models based on classical designs or threshold rules, which is fraught with false alarms or delays. Purpose: to propose and describe a practical approach to the use of contrastive self-supervised learning methods for early detection of anomalies in transport monitoring systems. Methods: contrastive learning methods with the InfoNCE loss function, lightweight convolutional encoders (1D-CNN/1D-ResNet), variational autoencoders (VAE) for reconstruction control, attention mechanisms for assessing sensor contributions, adaptive calibration of thresholds based on the exponential moving average (EMA), representation clustering for multiple normal modes, and testing scenarios on semi-real data are applied. Results: features of multisensory time series have been identified, limitations of field devices have been described, and ways to reduce false positives with limited markup have been identified. Algorithmic blocks for practical implementation are proposed: time series augmentation, adaptive threshold calibration, attention mechanisms for explainability and validation on semi-real data. A hybrid anomaly criterion based on contrastive and reconstruction scores has been developed. Practical significance: the introduction of a contrast unit with adaptive calibration provides a decrease in the time for detecting anomalies and a decrease in the frequency of false positives compared to basic VAE and threshold systems. Discussion: integration of contrast modules and adaptive calibration into existing transport monitoring systems using a three-level alarm system with per-channel explanation.

 контрастивное самообучение ранняя детекция мультисенсорный мониторинг мониторинг транспорта адаптивная калибровка contrastive self-supervised learning early detection multi-sensor monitoring transport monitoring adaptive calibration

Vehicle-as-a-Sensor Approach for Urban Track Anomaly Detection / V. Sruk [et al.] // Sensors. 2025. Vol. 25, iss. 21. Art. 6679. 24 p. DOI: 10.3390/s25216679 Vehicle-as-a-Sensor Approach for Urban Track Anomaly Detection / V. Sruk [et al.] // Sensors. 2025. Vol. 25, iss. 21. Art. 6679. 24 p. DOI: 10.3390/s25216679 Thill M., Konen W., Bäck T. Online Anomaly Detection on the Webscope S5 Dataset: A Comparative Study // Proceedings of the 2017 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS 2017) (Ljubljana, Slovenia, 31 May — 2 June 2017). Institute of Electrical and Electronics Engineers, 2017. 8 p. DOI: 10.1109/EAIS.2017.7954844 Thill M., Konen W., Bäck T. Online Anomaly Detection on the Webscope S5 Dataset: A Comparative Study // Proceedings of the 2017 IEEE Workshop on Evolving and Adaptive Intelligent Systems (EAIS 2017) (Ljubljana, Slovenia, 31 May — 2 June 2017). Institute of Electrical and Electronics Engineers, 2017. 8 p. DOI: 10.1109/EAIS.2017.7954844 Kingma D. P., Welling M. Auto-Encoding Variational Bayes // Proceedings of the Second International Conference on Learning Representations (ICLR 2014) (Banff, Canada, 14–16 April 2014). 14 p. DOI: 10.48550/arXiv.1312.6114 Kingma D. P., Welling M. Auto-Encoding Variational Bayes // Proceedings of the Second International Conference on Learning Representations (ICLR 2014) (Banff, Canada, 14–16 April 2014). 14 p. DOI: 10.48550/arXiv.1312.6114 Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding / K. Hundman [et al.] // Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘18) (London, United Kingdom, 19–23 August 2018). New York: Association for Computing Machinery, 2018. Pp. 387–395. DOI: 10.1145/3219819.3219845 Detecting Spacecraft Anomalies Using LSTMs and Nonparametric Dynamic Thresholding / K. Hundman [et al.] // Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘18) (London, United Kingdom, 19–23 August 2018). New York: Association for Computing Machinery, 2018. Pp. 387–395. DOI: 10.1145/3219819.3219845 A Simple Framework for Contrastive Learning of Visual Representations / T. Chen [et al.] // Proceedings of the 37th International Conference on Machine Learning (ICML 2020) (Virtual Event, 13–18 July 2020). Proceedings of Machine Learning Research. 2020. Vol. 119. Pp. 1597–1607. A Simple Framework for Contrastive Learning of Visual Representations / T. Chen [et al.] // Proceedings of the 37th International Conference on Machine Learning (ICML 2020) (Virtual Event, 13–18 July 2020). Proceedings of Machine Learning Research. 2020. Vol. 119. Pp. 1597–1607. USAD: Unsupervised Anomaly Detection on Multivariate Time Series / J. Audibert [et al.] // Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD‘20), (Virtual Event, 6–10 July 2020). New York: Association for Computing Machinery, 2020. Pp. 3395–3404. DOI: 10.1145/3394486.3403392 USAD: Unsupervised Anomaly Detection on Multivariate Time Series / J. Audibert [et al.] // Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD‘20), (Virtual Event, 6–10 July 2020). New York: Association for Computing Machinery, 2020. Pp. 3395–3404. DOI: 10.1145/3394486.3403392 Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection / B. Zong [et al.] // Proceedings of the Sixth International Conference on Learning Representations (ICLR 2018) (Vancouver, Canada, 30 April — 3 May 2018). 19 p. URL: http://openreview.net/forum?id=BJJLHbb0- (дата обращения: 09.05.2026). Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection / B. Zong [et al.] // Proceedings of the Sixth International Conference on Learning Representations (ICLR 2018) (Vancouver, Canada, 30 April — 3 May 2018). 19 p. URL: http://openreview.net/forum?id=BJJLHbb0- (data obrascheniya: 09.05.2026). Deng A., Hooi B. Graph Neural Network-Based Anomaly Detection in Multivariate Time Series // Proceedings of the 35th Conference on Artificial Intelligence (AAAI 2021) (Virtual Event, 2–9 February 2021). Proceedings of the AAAI Conference on Artificial Intelligence. 2021. Vol. 35, no. 5. Pp. 4027–4035. DOI: 10.1609/aaai.v35i5.16523 Deng A., Hooi B. Graph Neural Network-Based Anomaly Detection in Multivariate Time Series // Proceedings of the 35th Conference on Artificial Intelligence (AAAI 2021) (Virtual Event, 2–9 February 2021). Proceedings of the AAAI Conference on Artificial Intelligence. 2021. Vol. 35, no. 5. Pp. 4027–4035. DOI: 10.1609/aaai.v35i5.16523 On the Evaluation of Unsupervised Outlier Detection: Measures, Datasets, and an Empirical Study / G. O. Campos [et al.] // Data Mining and Knowledge Discovery. 2016. Vol. 30, iss. 4. Pp. 891–927. DOI: 10.1007/s10618-015-0444-8 On the Evaluation of Unsupervised Outlier Detection: Measures, Datasets, and an Empirical Study / G. O. Campos [et al.] // Data Mining and Knowledge Discovery. 2016. Vol. 30, iss. 4. Pp. 891–927. DOI: 10.1007/s10618-015-0444-8 Federated Learning: Challenges, Methods, and Future Directions / T. Li [et al.] // IEEE Signal Processing Magazine. 2020. Vol. 37, no. 3. Pp. 50–60. DOI: 10.1109/MSP.2020.2975749 Federated Learning: Challenges, Methods, and Future Directions / T. Li [et al.] // IEEE Signal Processing Magazine. 2020. Vol. 37, no. 3. Pp. 50–60. DOI: 10.1109/MSP.2020.2975749 Schneider J. Wenig P., Papenbrock T. Distributed Detection of Sequential Anomalies in Univariate Time Series // The International Journal on Very Large Data Bases. 2021. Vol. 30, iss. 4. Pp. 579–602. DOI: 10.1007/s00778-021-00657-6 Schneider J. Wenig P., Papenbrock T. Distributed Detection of Sequential Anomalies in Univariate Time Series // The International Journal on Very Large Data Bases. 2021. Vol. 30, iss. 4. Pp. 579–602. DOI: 10.1007/s00778-021-00657-6 Long Short Term Memory Networks for Anomaly Detection in Time Series / P. Malhotra [et al.] // Proceedings of the 23rd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2015) (Bruges, Belgium, 22–24 April 2015). i6doc.com Publishing, 2015. Pp. 89–94. Long Short Term Memory Networks for Anomaly Detection in Time Series / P. Malhotra [et al.] // Proceedings of the 23rd European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2015) (Bruges, Belgium, 22–24 April 2015). i6doc.com Publishing, 2015. Pp. 89–94. Robust Anomaly Detection for Multivariate Time Series Through Stochastic Recurrent Neural Network / Y. Su [et al.] // Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘19) (Anchorage, AK, USA, 4–8 August 2019). New York: Association for Computing Machinery, 2019. Pp. 2828–2837. DOI: 10.1145/3292500.3330672 Robust Anomaly Detection for Multivariate Time Series Through Stochastic Recurrent Neural Network / Y. Su [et al.] // Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD ‘19) (Anchorage, AK, USA, 4–8 August 2019). New York: Association for Computing Machinery, 2019. Pp. 2828–2837. DOI: 10.1145/3292500.3330672