|
Аннотация.
Целью исследования является разработка и валидация метода обнаружения скрытых неисправностей в системах охлаждения зданий на основе включения топологической схемы в архитектуру графовой рекуррентной нейронной сети GConvGRU, способной учитывать как топологические, так и временные зависимости между сенсорами HVAC-систем. В работе проведено сравнение предложенного подхода с классической LSTM-архитектурой на трёх независимых открытых датасетах - экспериментальный стенд LBNL, реальные данные офисного здания и симулированная модель многоэтажного офиса, охватывающих реальные и симулированные условия эксплуатации. Результаты демонстрируют стабильное превосходство GConvGRU по метрикам F1-score и Recall при сохранении высокой точности, подтверждая, что интеграция физической структуры системы в виде ориентированного графа существенно повышает способность модели выявлять постепенные аномалии на фоне нестационарных рабочих режимов.
Ключевые слова:
центр обработки данных, временные ряды, поиск аномалий, графовые нейронные сети, утечка хладагента.
DOI 10.14357/20718632260311
EDN EOQPZW
Стр. 122-135.
Литература
1. Pérez-Lombard L., Ortiz J., Coronel J.F., Maestre I.R. A review of HVAC systems requirements in building energy regulations // Energy and Buildings. 2011. Vol. 43. No. 2-3. P. 255-268. https://doi.org/10.1016/j.enbuild.2010.10.025 2. Ni J., Bai X. A review of air conditioning energy performance in data centers // Renewable and Sustainable Energy Reviews. 2017. Vol. 67. P. 625-640. https://doi.org/10.1016/j.rser.2016.09.050 3. Li Y., O'Neill Z. A critical review of fault modeling of HVAC systems in buildings // Building Simulation. 2018. Vol. 11. No. 5. P. 953-975. https://doi.org/10.1007/s12273-018-0458-4 4. Zhang R., Hong T. Modeling of HVAC operational faults in building performance simulation // Applied Energy. 2017. Vol. 202. P. 178-188. https://doi.org/10.1016/j.apenergy.2017.05.153 5. Meng L., Yi R., Finocchiaro W., Alemu A. Integrating Building Management Systems (BMS) and Building Information Model (BIM) to improve HVAC efficiency: a case study in Darwin // 2023 6th International Conference on E-Business, Information Management and Computer Science (EBIMCS): proceedings, Darwin, 15-17 December 2023. New York: ACM, 2023. P. 41-46. https://doi.org/10.1145/3644479.3644494 6. Levy M., Hallstrom J.O. A new approach to data center infrastructure monitoring and management (DCIMM) // 2017 IEEE 7th Annual Computing and Communication Workshop and Conference (CCWC), Las Vegas, 9-11 January 2017. IEEE, 2017. P. 1-6. https://doi.org/10.1109/CCWC.2017.7868412 7. Asgari S., Gupta R., Puri I.K., Zheng R. A data-driven approach to simultaneous fault detection and diagnosis in data centers // Applied Soft Computing. 2021. Vol. 110. Art. 107638. https://doi.org/10.1016/j.asoc.2021.107638 8. von Rhein J., Henze G.P., Long N., Fu Y. Development of a topology analysis tool for fifth-generation district heating and cooling networks // Energy Conversion and Management. 2019. Vol. 196. P. 705-716. https://doi.org/10.1016/j.enconman.2019.05.066 9. Jiang J.-R., Kao J.-B., Li Y.-L. Semi-supervised time series anomaly detection based on statistics and deep learning // Applied Sciences. 2021. Vol. 11. No. 15. Art. 6698. https://doi.org/10.3390/app11156698 10. Yin C., Zhang S., Wang J., Xiong N.N. Anomaly detection based on convolutional recurrent autoencoder for IoT time series // IEEE Transactions on Systems, Man, and Cybernetics: Systems. 2022. Vol. 52. No. 1. P. 112-122. https://doi.org/10.1109/TSMC.2020.2968516 11. Wen Q., Zhou T., Zhang C., Chen W., Ma Z., Yan J., Sun L. Transformers in time series: a survey // arXiv preprint. 2023. arXiv:2202.07125. URL: https://arxiv.org/abs/2202.07125 (дата обращения: 19.12.2025). 12. Yu Y., Si X., Hu C., Zhang J. A review of recurrent neural networks: LSTM cells and network architectures // Neural Computation. 2019. Vol. 31. No. 7. P. 1235-1270. https://doi.org/10.1162/neco_a_01199 13. Nosouhian S., Nosouhian F., Kazemi Khoshouei A. A review of recurrent neural network architecture for sequence learning: comparison between LSTM and GRU // Preprints. 2021. https://doi.org/10.20944/preprints202107.0252.v1 14. Shumway R.H., Stoffer D.S. ARIMA models // Time Series Analysis and Its Applications: With R Examples. 4th ed. Cham: Springer, 2017. P. 75-163. (Springer Texts in Statistics). https://doi.org/10.1007/978-3-319-52452-8_3 15. Khemani B., Patil S., Kotecha K., Tanwar S. A review of graph neural networks: concepts, architectures, techniques, challenges, datasets, applications, and future directions // Journal of Big Data. 2024. Vol. 11. No. 1. Art. 18. https://doi.org/10.1186/s40537-023-00876-4 16. Li S., Xu L., Jing Y., Yin H., Li X., Guan X. High-quality vegetation index product generation: a review of NDVI time series reconstruction techniques // International Journal of Applied Earth Observation and Geoinformation. 2021. Vol. 105. Art. 102640. https://doi.org/10.1016/j.jag.2021.102640 17. Villa-Pérez M.E., Álvarez-Carmona M.Á., Loyola-González O., Medina-Pérez M.A., Velazco-Rossell J.C., Choo K.-K.R. Semi-supervised anomaly detection algorithms: a comparative summary and future research directions // Knowledge-Based Systems. 2021. Vol. 218. Art. 106878. https://doi.org/10.1016/j.knosys.2021.106878 18. Blázquez-García A., Conde A., Mori U., Lozano J.A. A review on outlier/anomaly detection in time series data // 19. Seo Y., Defferrard M., Vandergheynst P., Bresson X. Structured sequence modeling with graph convolutional recurrent networks // Neural Information Processing: ICONIP 2018, Siem Reap, 13-16 December 2018. Cham: Springer, 2018. P. 362-373. (Lecture Notes in Computer Science; vol. 11301). https://doi.org/10.1007/978-3-030-04167-0_33 20. Bergmeir C., Benitez J.M. On the use of cross-validation for time series predictor evaluation // Information Sciences. 2012. Vol. 191. P. 192-213. https://doi.org/10.1016/j.ins.2011.12.028 21. Thara D.K., Premasudha B.G., Xiong F. Auto-detection of epileptic seizure events using deep neural network with different feature scaling techniques // Pattern Recognition Letters. 2019. Vol. 128. P. 544-550. https://doi.org/10.1016/j.patrec.2019.10.029 22. Kingma D.P., Ba J. Adam: a method for stochastic optimization // arXiv: preprint. 2014. arXiv:1412.6980. URL: https://arxiv.org/abs/1412.6980 (data obrащения: 19.12.2025). 23. Chicco D., Warrens M.J., Jurman G. The coefficient of determination R-squared is more informative than SMAPE, MAE, MAPE, MSE and RMSE in regression analysis evaluation // PeerJ Computer Science. 2021. Vol. 7. Art. e623. https://doi.org/10.7717/peerj-cs.623 24. Wei L., Keogh E. Semi-supervised time series classification // Proceedings of the 12th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Philadelphia, 20-23 August 2006. New York: ACM, 2006. P. 748-753. https://doi.org/10.1145/1150402.1150498 25. Granderson J., Lin G., Harding A., Im P., Chen Y. Dataset for building fault detection and diagnostics algorithm creation and performance testing: dataset. figshare, 2020. https://doi.org/10.6084/m9.figshare.11752740.v3 (дата обращения: 19.12.2025). 26. Çiftpınar A.B., Kanar P., Erzurum Çiçek Z.İ. Failure prediction using ensemble learning: a comparative study with synthetic and real-world datasets // Afyon Kocatepe University Journal of Science and Engineering. 2025. Vol. 25. P. 785-797. https://doi.org/10.35414/akufenubid.1571811 27. Wang S. Labelled dataset for FDD in different building types: auditorium, hospital, and large-scale office: dataset. figshare, 2025. https://doi.org/10.6084/m9.figshare.27147678.v3 (дата обращения: 19.12.2025). 28. Wang S., Eum I., Park S., Kim J. A semi-labelled dataset for fault detection in air handling units from a large-scale office // Data in Brief. 2024. Vol. 57. Art. 110956. https://doi.org/10.1016/j.dib.2024.110956 29. Chen Y., Granderson J. LBNL FDD data sets: collection. figshare, 2023. https://doi.org/10.6084/m9.figshare.c.6486349.v1 (дата обращения: 19.12.2025).
|
