<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Publishing DTD v1.3 20210610//EN" "JATS-journalpublishing1-3.dtd">
<article article-type="research-article" dtd-version="1.3" xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink" xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance" xml:lang="ru"><front><journal-meta><journal-id journal-id-type="publisher-id">energsecurity</journal-id><journal-title-group><journal-title xml:lang="ru">Надежность и безопасность энергетики</journal-title><trans-title-group xml:lang="en"><trans-title>Safety and Reliability of Power Industry</trans-title></trans-title-group></journal-title-group><issn pub-type="ppub">1999-5555</issn><issn pub-type="epub">2542-2057</issn><publisher><publisher-name>ООО «НПО Энергобезопасность»</publisher-name></publisher></journal-meta><article-meta><article-id pub-id-type="doi">10.24223/1999-5555-2026-19-2-152-158</article-id><article-id custom-type="elpub" pub-id-type="custom">energsecurity-1094</article-id><article-categories><subj-group subj-group-type="heading"><subject>Research Article</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="ru"><subject>ОБЩИЕ ВОПРОСЫ НАДЕЖНОСТИ И БЕЗОПАСНОСТИ ЭНЕРГЕТИКИ</subject></subj-group><subj-group subj-group-type="section-heading" xml:lang="en"><subject>GENERAL ISSUES RELATED TO RELIABILITY AND SAFETY OF THE POWER INDUSTRY</subject></subj-group></article-categories><title-group><article-title>Гибридный подход нейронных сетей для предотвращения аварий и утечек в трубопроводах и котельных по видеопотоку</article-title><trans-title-group xml:lang="en"><trans-title>A hybrid neural network approach for preventing accidents and leaks in pipelines and boiler rooms using video stream analysis</trans-title></trans-title-group></title-group><contrib-group><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Каширин</surname><given-names>М. В.</given-names></name><name name-style="western" xml:lang="en"><surname>Kashirin</surname><given-names>M. V.</given-names></name></name-alternatives><bio xml:lang="ru"><p>ул. Авиамоторная, 8а, 111024, г. Москва</p><p>Адрес для переписки:</p><p>Каширин М. В., АО «Мытищинская теплосеть», ул. Колпакова, 20, 141008, г. Мытищи</p></bio><bio xml:lang="en"><p>8a Aviamotornaya str., 111024, Moscow</p><p>Address for correspondence:</p><p>Kashirin M. V., JSC “Mytishchi Heating Network”, Kolpakova str., 20, 141008, Mytishchi</p></bio><email xlink:type="simple">maksimus_kashiri@mail.ru</email><xref ref-type="aff" rid="aff-1"/></contrib><contrib contrib-type="author" corresp="yes"><name-alternatives><name name-style="eastern" xml:lang="ru"><surname>Кравченко</surname><given-names>В. Н.</given-names></name><name name-style="western" xml:lang="en"><surname>Kravchenko</surname><given-names>V. N.</given-names></name></name-alternatives><bio xml:lang="ru"><p>ул. Авиамоторная, 8а, 111024, г. Москва</p></bio><bio xml:lang="en"><p>8a Aviamotornaya str., 111024, Moscow</p></bio><xref ref-type="aff" rid="aff-1"/></contrib></contrib-group><aff-alternatives id="aff-1"><aff xml:lang="ru"><institution>Ордена Трудового Красного Знамени ФГБОУ ВО «Московский технический университет связи и информатики»</institution><country>Россия</country></aff><aff xml:lang="en"><institution>Moscow Technical University of Communications and Informatics</institution><country>Russian Federation</country></aff></aff-alternatives><pub-date pub-type="collection"><year>2026</year></pub-date><pub-date pub-type="epub"><day>02</day><month>08</month><year>2026</year></pub-date><volume>19</volume><issue>2</issue><fpage>152</fpage><lpage>158</lpage><permissions><copyright-statement>Copyright &amp;#x00A9; Каширин М.В., Кравченко В.Н., 2026</copyright-statement><copyright-year>2026</copyright-year><copyright-holder xml:lang="ru">Каширин М.В., Кравченко В.Н.</copyright-holder><copyright-holder xml:lang="en">Kashirin M.V., Kravchenko V.N.</copyright-holder><license xml:lang="ru" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>Данная работа распространяется под лицензией Creative Commons Attribution 4.0.</license-p></license><license xml:lang="en" license-type="creative-commons-attribution" xlink:href="https://creativecommons.org/licenses/by/4.0/" xlink:type="simple"><license-p>This work is licensed under a Creative Commons Attribution 4.0 License.</license-p></license></permissions><self-uri xlink:href="https://www.sigma08.ru/jour/article/view/1094">https://www.sigma08.ru/jour/article/view/1094</self-uri><abstract><p>Рассматривается комплексный подход к решению проблемы аварий и утечек в трубопроводных и котельных системах, сочетающий методы прогнозирования рисков и современные алгоритмы компьютерного зрения для анализа видеопотока. Актуальность исследования обусловлена высокой аварийностью инженерных сетей, значительными экономическими потерями и экологическими рисками, связанными с несвоевременным обнаружением утечек. Целью работы является разработка и экспериментальная оценка гибридного метода предотвращения аварий и утечек на основе прогнозирования рисков и нейросетевого анализа видеоданных в реальном времени. В рамках исследования проанализированы архитектуры сверточных нейронных сетей, применяемые для обработки видеопотока, и показаны ограничения традиционных методов мониторинга, основанных на точечных датчиках давления и расхода. На основе исторических, эксплуатационных и телеметрических данных формируется прогнозная таблица рисков, определяющая приоритетность контроля потенциально опасных участков инженерных систем. Для участков с повышенным уровнем риска применяется гибридная нейросетевая модель, включающая модуль пространственно-временной детекции аномалий и модуль семантической сегментации визуальных признаков утечек. Результаты экспериментального исследования, проведённого на наборе данных объемом 120 часов видеозаписей с 20 промышленных объектов, демонстрируют точность обнаружения до 94,7%, F1-score 92,9% и среднее время отклика около 22 секунд. В заключении отмечается, что предложенное решение позволяет перейти от реактивной к проактивной модели эксплуатации инженерных систем, повысить уровень промышленной и экологической безопасности и может быть интегрировано с существующими средствами контроля для расширения зоны мониторинга.</p></abstract><trans-abstract xml:lang="en"><p>A comprehensive approach to solving the problem of accidents and leaks in pipeline and boiler systems is considered, combining risk forecasting methods with modern computer vision algorithms for video stream analysis. The relevance of the study is due to the high accident rate of engineering networks, significant economic losses, and environmental risks associated with delayed leak detection. The aim of the work is to develop and experimentally evaluate a hybrid method for accident and leak prevention based on risk forecasting and real-time neural network analysis of video data. Within the framework of the study, convolutional neural network architectures used for video stream processing are analyzed, and the limitations of traditional monitoring methods based on point pressure and flow sensors are shown. Based on historical, operational, and telemetry data, a predictive risk table is formed determining the priority of monitoring potentially hazardous sections of engineering systems. For sections with elevated risk levels, a hybrid neural network model is applied, including a spatiotemporal anomaly detection module and a semantic segmentation module for identifying visual indicators of leaks. The results of the experimental study, conducted on a dataset of 120 hours of video recordings from 20 industrial facilities, demonstrate detection accuracy of up to 94.7%, an F1-score of 92.9%, and an average response time of approximately 22 seconds. In conclusion it is noted that the proposed solution allows a transition from a reactive to a proactive model of engineering systems, enhances industrial and environmental safety, and can be integrated with existing monitoring means to expand the monitoring zone.</p></trans-abstract><kwd-group xml:lang="ru"><kwd>нейронные сети</kwd><kwd>компьютерное зрение</kwd><kwd>детекция утечек</kwd><kwd>трубопроводы и котельные системы</kwd><kwd>видеопоток</kwd><kwd>мониторинг</kwd><kwd>прогнозирование рисков</kwd><kwd>семантическая сегментация</kwd></kwd-group><kwd-group xml:lang="en"><kwd>neural networks</kwd><kwd>computer vision</kwd><kwd>leak detection</kwd><kwd>pipeline and boiler systems</kwd><kwd>video stream</kwd><kwd>monitoring</kwd><kwd>risk forecasting</kwd><kwd>semantic segmentation</kwd></kwd-group></article-meta></front><back><ref-list><title>References</title><ref id="cit1"><label>1</label><citation-alternatives><mixed-citation xml:lang="ru">Review and analysis of pipeline leak detection methods. N. V. S. Korlapati, F. Khan, Q. Noor, S. Mirza, S. Vaddiraju. Journal of Pipeline Science and Engineering 2022; 2 (4): 100074.</mixed-citation><mixed-citation xml:lang="en">Review and analysis of pipeline leak detection methods. N. V. S. Korlapati, F. Khan, Q. Noor, S. Mirza, S. Vaddiraju. Journal of Pipeline Science and Engineering 2022; 2 (4): 100074. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit2"><label>2</label><citation-alternatives><mixed-citation xml:lang="ru">Нгула Б.-Ж., Кузяков О. Н. Обнаружение утечек в нефте- и газопроводах с помощью интернет вещей и алгоритма глубокого об-учения. Научный аспект 2024; (5).</mixed-citation><mixed-citation xml:lang="en">Ngula B.-J., Kuzyakov O. N. Leak Detection in Oil and Gas Pipelines Using the Internet of Things and Deep Learning Algorithms. Scientific Aspect 2024; (5). (In Russ.)</mixed-citation></citation-alternatives></ref><ref id="cit3"><label>3</label><citation-alternatives><mixed-citation xml:lang="ru">Пленкина К. О. Применение методов визуализации утечек газа. Молодой ученый 2025; 21 (572): 94 – 96.</mixed-citation><mixed-citation xml:lang="en">Plenkina K. O. Application of Gas Leak Visualization Methods. Young Scientist 2025; 21 (572): 94 – 96. (In Russ.)</mixed-citation></citation-alternatives></ref><ref id="cit4"><label>4</label><citation-alternatives><mixed-citation xml:lang="ru">Evaluation of deep learning approaches for oil &amp; gas pipeline leak detection using wireless sensor networks. C. Spandonidis, P. Theodoropoulos, F. Giannopoulos, N. Galiatsatos, A. Petsa. Engineering Applications of Artificial Intelligence 2022; 114: 104890.</mixed-citation><mixed-citation xml:lang="en">Evaluation of deep learning approaches for oil &amp; gas pipeline leak detection using wireless sensor networks. C. Spandonidis, P. Theodoropoulos, F. Giannopoulos, N. Galiatsatos, A. Petsa. Engineering Applications of Artificial Intelligence 2022; 114: 104890. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit5"><label>5</label><citation-alternatives><mixed-citation xml:lang="ru">Saleem F., Ahmad Z., Kim J.-M. Real-Time Pipeline Leak Detection: A Hybrid Deep Learning Approach Using Acoustic Emission Signals. Applied Sciences 2025; 15 (1): 185.</mixed-citation><mixed-citation xml:lang="en">Saleem F., Ahmad Z., Kim J.-M. Real-Time Pipeline Leak Detection: A Hybrid Deep Learning Approach Using Acoustic Emission Signals. Applied Sciences 2025; 15 (1): 185. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit6"><label>6</label><citation-alternatives><mixed-citation xml:lang="ru">Automated leakage detection method of pipeline networks under complicated backgrounds by combining infrared thermography and Faster R-CNN technique. J. Xie, Y. Zhang, Z. He, P. Liu, Y. Qin, Z. Wang, C. Xu. Process Safety and Environmental Protection 2023; 174: 39 – 52.</mixed-citation><mixed-citation xml:lang="en">Automated leakage detection method of pipeline networks under complicated backgrounds by combining infrared thermography and Faster R-CNN technique. J. Xie, Y. Zhang, Z. He, P. Liu, Y. Qin, Z. Wang, C. Xu. Process Safety and Environmental Protection 2023; 174: 39 – 52. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit7"><label>7</label><citation-alternatives><mixed-citation xml:lang="ru">Video Swin Transformer. Z. Liu, J. Ning, Y. Cao, Y. Wei, Z. Zhang, S. Lin, H. Hu. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022: 3202 – 3211.</mixed-citation><mixed-citation xml:lang="en">Video Swin Transformer. Z. Liu, J. Ning, Y. Cao, Y. Wei, Z. Zhang, S. Lin, H. Hu. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) 2022: 3202 – 3211. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit8"><label>8</label><citation-alternatives><mixed-citation xml:lang="ru">Wang J., Ruhaiyem N. I. R., Fu P. A Comprehensive Review of U-Net and Its Variants: Advances and Applications in Medical Image Segmentation. IET Image Processing 2025.10.1049/ipr2.70019.</mixed-citation><mixed-citation xml:lang="en">Wang J., Ruhaiyem N. I. R., Fu P. A Comprehensive Review of U-Net and Its Variants: Advances and Applications in Medical Image Segmentation. IET Image Processing 2025.10.1049/ipr2.70019. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit9"><label>9</label><citation-alternatives><mixed-citation xml:lang="ru">Khalid S., Azad M. M., Kim H. S. Real-world steam powerplant boiler tube leakage detection using hybrid deep learning. Mathematics 2024; 12 (24): 3887.</mixed-citation><mixed-citation xml:lang="en">Khalid S., Azad M. M., Kim H. S. Real-world steam powerplant boiler tube leakage detection using hybrid deep learning. Mathematics 2024; 12 (24): 3887. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit10"><label>10</label><citation-alternatives><mixed-citation xml:lang="ru">Aragonés R., Oliver J., Ferrer C. Thermoelectric Generator-Powered Long-Range Wireless Real-Time Steam Leak Detection in Steam Traps. Future Internet 2024; 16 (12): 474.</mixed-citation><mixed-citation xml:lang="en">Aragonés R., Oliver J., Ferrer C. Thermoelectric Generator-Powered Long-Range Wireless Real-Time Steam Leak Detection in Steam Traps. Future Internet 2024; 16 (12): 474. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit11"><label>11</label><citation-alternatives><mixed-citation xml:lang="ru">Pipeline leakage identification method based on DPR-net and distributed optical fiber acoustic sensing technology. Y. Zhan, L. Liu, Z. Wang, W. Zhang, K. Li, Y. Liu, Y. Liu, J. Wu, C. Yici, B. Chen, Q. Ye, Q. Ye, H. Cai. Optics Communications 2025; 574: 131096.</mixed-citation><mixed-citation xml:lang="en">Pipeline leakage identification method based on DPR-net and distributed optical fiber acoustic sensing technology. Y. Zhan, L. Liu, Z. Wang, W. Zhang, K. Li, Y. Liu, Y. Liu, J. Wu, C. Yici, B. Chen, Q. Ye, Q. Ye, H. Cai. Optics Communications 2025; 574: 131096. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit12"><label>12</label><citation-alternatives><mixed-citation xml:lang="ru">Leak detection in pipelines based on acoustic emission and growing neural gas network utilizing unlabeled healthy condition data. A. Mishra, J. Dhebar, B. Das, S. S. Patel, A. Rai. Flow Measurement and Instrumentation 2025; 102: 102816.</mixed-citation><mixed-citation xml:lang="en">Leak detection in pipelines based on acoustic emission and growing neural gas network utilizing unlabeled healthy condition data. A. Mishra, J. Dhebar, B. Das, S. S. Patel, A. Rai. Flow Measurement and Instrumentation 2025; 102: 102816. (In Eng.)</mixed-citation></citation-alternatives></ref><ref id="cit13"><label>13</label><citation-alternatives><mixed-citation xml:lang="ru">Лозовой С. В. Системы обнаружения утечек на нефтяных и газовых месторождениях. Отходы и ресурсы 2024; 11 (1).</mixed-citation><mixed-citation xml:lang="en">Lozovoy S. V. Leak Detection Systems at Oil and Gas Fields. Waste and Resources 2024; 11(1). (In Russ.)</mixed-citation></citation-alternatives></ref></ref-list><fn-group><fn fn-type="conflict"><p>The authors declare that there are no conflicts of interest present.</p></fn></fn-group></back></article>
