ipolytechiPolytech JournaliPolytech Journal2782-40042782-6341Irkutsk National Research Technical University10.21285/1814-3520-2020-5-1093-1104ipolytech-439Research ArticleЭНЕРГЕТИКАPOWER ENGINEERINGМашинное обучение как инструмент повышения эффективности управления жизненным циклом высоковольтного электрооборудованияMachine learning as a tool of high-voltage electrical equipment lifecycle control enhancementХальясмааА. И.KhalyasmaaA. I.

Хальясмаа Александра Ильмаровна, кандидат технических наук, доцент, доцент кафедры «Электрические станции»

630073, г. Новосибирск, пр. Карла Маркса, 20

Alexandra I. Khalyasmaa, Cand. Sci. (Eng.), Associate Professor of the Department of Electric Power Stations

20, Karl Marx Ave., Novosibirsk 630073

lkhalyasmaa@mail.ruНовосибирский государственный технический университетРоссияNovosibirsk State Technical UniversityRussian Federation20201311202024510931104Copyright © Хальясмаа А.И., 20202020Хальясмаа А.И.Khalyasmaa A.I.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://ipolytech.elpub.ru/jour/article/view/439

Цель – опытный анализ практической реализации подсистемы оценки технического состояния высоковольтного электрооборудования в рамках решения задачи управления его жизненным циклом на основе методов машинного обучения с учетом анализа влияния режимов работы внешней электроэнергетической системы. Для решения задачи анализа технического состояния оборудования – распознавания образов состояния оборудования – использовался градиентный бустинг XGBoost (XGB) на основе решающих деревьев, основными преимуществами которого являются способность обработки данных с пропусками и эффективность работы с табличными данными для решения задач классификации и регрессии. Предложено описание формирования корректной и достаточной структуры исходной базы данных для распознавания образа состояния высоковольтного оборудования на основе данных его технического диагностирования и алгоритма формирования обучающих и тестовых выборок для повышения точности идентификации фактического состояния оборудования, а также описание и обоснование применения метода машинного обучения и соответствующих метрик ошибок классификации состояний. На основе анализа фактического состояния силовых трансформаторов и выключателей сформированы перечни параметров технического диагностирования, оказывающие наибольшее влияние на точность идентификации состояний, а также доказана эффективность применения режимных параметров в качестве дополнительных признаков. Установлено, что учет режимных параметров, полученных расчетных путем, в составе обучающей выборки для идентификации состояния высоковольтного оборудования дает возможность повысить точность настройки. Разработанные структура и подходы к анализу технического состояния оборудования с использованием схемно-режимной информации наряду с диагностическими данными обеспечивают информационную связь задач технологического и оперативно-диспетчерского управления, что позволяет рассматривать задачу ведения электрических режимов энергосистем с позиции технического состояния электросетевого оборудования и выявлять наиболее приоритетные задачи эксплуатационного обслуживания для снятия сетевых и системных ограничений.

The purpose of the study is to analyze the practical implementation of high-voltage electrical equipment technical state estimation subsystems as a part of solving the lifecycle management problem based on machine learning methods and taking into account the effect of the adjacent power system operation modes. To deal with the problem of power equipment technical state analysis, i.e. power equipment state pattern recognition, XGBoost based on gradient boosting decision tree algorithm is used. Its main advantages are the ability to process gapped data and efficient operation with tabular data for solving classification and regression problems. The author suggests the formation procedure of correct and sufficient initial database for high-voltage equipment state pattern recognition based on its technical diagnostic data and the algorithm for training and testing sets creation in order to improve the identification accuracy of power equipment actual state. The description and justification of the machine learning method and corresponding error metrics are also provided. Based on the actual states of power transformers and circuit breakers the sets of technical diagnostic parameters that have the greatest impact on the accuracy of state identification are formed. The effectiveness of using power systems operation parameters as additional features is also confirmed. It is determined that the consideration of operation parameters obtained by calculation as a part of the training set for high-voltage equipment technical state identification makes it possible to improve the tuning accuracy. The developed structure and approaches to power equipment technical state analysis supplemented by power system operation mode data and diagnostic results provide an information link between the tasks of technological and dispatch control. This allows us to consider the task of power system operation mode planning from the standpoint of power equipment technical state and identify the priorities in repair and maintenance to eliminate power network “bottlenecks”.

жизненный циклвысоковольтное оборудованиеоценка технического состояниямашинное обучениережимные параметрыlife cyclehigh-voltage equipmenttechnical state estimationmachine learningoperation parametersReferencesCai Zhongyi, Wang Zezhou, Chen Yunxiang, Guo Jiansheng, Xiang Huachun. Remaining useful lifetime prediction for equipment based on nonlinear implicit degradation modeling // Journal of Systems Engineering and Electronics. 2020. Vol. 31. Issue 1. P. 194–205. https://doi.org/10.21629/JSEE.2020.01.19Cai Zhongyi, Wang Zezhou, Chen Yunxiang, Guo Jiansheng, Xiang Huachun. Remaining useful lifetime prediction for equipment based on nonlinear implicit degradation modeling. Journal of Systems Engineering and Electronics. 2020;31(1):194–205. https://doi.org/10.21629/JSEE.2020.01.19Cao Wen, He Baosheng, Shen Wei, Shi Xuetao, Wu Kai, Tatsuki Okamoto, еt al. An approach for economic assessment on oil-paper insulation diagnosis through accelerated aging experiments // IEEE Transactions on Dielectrics and Electrical Insulation. 2014. Vol. 21. Issue 4. P. 1842–1850. https://doi.org/10.1109/TDEI.2014.004370Cao Wen, He Baosheng, Shen Wei, Shi Xuetao, Wu Kai, Tatsuki Okamoto, еt al. An approach for economic assessment on oil-paper insulation diagnosis through accelerated aging experiments. IEEE Transactions on Dielectrics and Electrical Insulation. 2014;21(4):1842–1850. https://doi.org/10.1109/TDEI.2014.004370Ittiphong Yongyee, Cattareeya Suwanasri, Thanapong Suwanasri, Waraporn Luejai. Condition assessment in transmission line asset for maintenance management // International Electrical Engineering Congress. 2018. https://doi.org/10.1109/IEECON.2018.8712276Ittiphong Yongyee, Cattareeya Suwanasri, Thanapong Suwanasri, Waraporn Luejai. Condition assessment in transmission line asset for maintenance management. In: International Electrical Engineering Congress. 2018. https://doi.org/10.1109/IEECON.2018.8712276Li Shengtao, Li Jianying. Condition monitoring and diagnosis of power equipment: review and prospective // High Voltage. 2017. Vol. 2. Issue 2. P. 82–91. https://doi.org/10.1049/hve.2017.0026Li Shengtao, Li Jianying. Condition monitoring and diagnosis of power equipment: review and prospective. High Voltage. 2017;2(2):82–91. https://doi.org/10.1049/hve.2017.0026Liu Yingyi, Xv Jiahuan, Yuan Haiwen, Lv Jianxun, Ma Zhao. Health assessment and prediction of overhead line based on health index // IEEE Transactions on Industrial Electronics. 2019. Vol. 66. No. 7. P. 5546–5557. https://doi.org/10.1109/TIE.2018.2868028Liu Yingyi, Xv Jiahuan, Yuan Haiwen, Lv Jianxun, Ma Zhao. Health assessment and prediction of overhead line based on health index. IEEE Transactions on Industrial Electronics. 2019;66(7):5546–5557. https://doi.org/10.1109/TIE.2018.2868028Xu Gao Huan, Ye Jun Xiang. Engine fault diagnosis based on support vector machine and noise analysis // Advanced Materials Research. 2012. Vol. 546–547. P. 97–101. https://doi.org/10.4028/www.scientific.net/amr.546-547.97Xu Gao Huan, Ye Jun Xiang. Engine fault diagnosis based on support vector machine and noise analysis. Advanced Materials Research. 2012;546–547:97–101. https://doi.org/10.4028/www.scientific.net/amr.546-547.97Ben Ali J., Chebel-Morello B., Saidi L., Malinowski S., Fnaiech F. Accurate bearing remaining useful life prediction based on weibull distribution and artificial neural network // Mechanical Systems and Signal Processing. 2015. Vol. 56–57. P. 150–172. https://doi.org/10.1016/j.ymssp.2014.10.014Ben Ali J, Chebel-Morello B, Saidi L, Malinowski S, Fnaiech F. Accurate bearing remaining useful life prediction based on weibull distribution and artificial neural network. Mechanical Systems and Signal Processing. 2015;56–57:150–172. https://doi.org/10.1016/j.ymssp.2014.10.014Ghasemi A., Yacout S., Ouali M.S. Evaluating the reliability function and the mean residual life for equipment with unobservable states // IEEE Transactions on Reliability. 2010. Vol. 59. Issue 1. P. 45–54. https://doi.org/10.1109/TR.2009.2034947Ghasemi A, Yacout S, Ouali MS. Evaluating the reliability function and the mean residual life for equipment with unobservable states. IEEE Transactions on Reliability. 2010;59(1):45–54. https://doi.org/10.1109/TR.2009.2034947Li Shengtao, Li Jianying. Condition monitoring and diagnosis of power equipment: review and prospective // High Voltage. 2017. Vol. 2. No. 2. P. 82–91. 6 https://doi.org/10.1049/hve.2017.0026Li Shengtao, Li Jianying. Condition monitoring and diagnosis of power equipment: review and prospective. High Voltage. 2017;2(2):82–91. https://doi.org/10.1049/hve.2017.0026Tanasescu G., Gorgan B., Busoi S., Badita A., Notingher P.V. Monitoring and diagnosis of electrical equipment using a web software. health index and remaining lifetime estimation // Conference on Diagnostics in Electrical Engineering (Diagnostika). 2016. https://doi.org/10.1109/DIAGNOSTIKA.2016.7736495Tanasescu G, Gorgan B, Busoi S, Badita A, Notingher PV. Monitoring and diagnosis of electrical equipment using a web software. Health Index and Remaining Lifetime Estimation. In: Conference on Diagnostics in Electrical Engineering (Diagnostika). 2016. https://doi.org/10.1109/DIAGNOSTIKA.2016.7736495Zhang Xiaoqing, Gockenbach E., Chen H.B., Liu Z.L., Yang L.H., Gao Kunlun, et al. Life management of 550kV SF6 circuit breakers // International Conference on High Voltage Engineering and Application. 2010. https://doi.org/10.1109/ICHVE.2010.5640763Zhang Xiaoqing, Gockenbach E, Chen HB, Liu ZL, Yang LH, Gao Kunlun, et al. Life management of 550kV SF6 circuit breakers. In: International Conference on High Voltage Engineering and Application. 2010. https://doi.org/10.1109/ICHVE.2010.5640763Kai Huang, Shuqiang Zhao. Evaluation of residual service life of high voltage circuit breaker // International Conference on Electricity Distribution. 2012. https://doi.org/10.1109/CICED.2012.6508698Kai Huang, Shuqiang Zhao. Evaluation of residual service life of high voltage circuit breaker. In: International Conference on Electricity Distribution. 2012. https://doi.org/10.1109/CICED.2012.6508698Asiri Ya., Vouk A., Renfort L., Clark D., Copper J. Neural network based classification of partial discharge in hv motors // Electrical Insulation Conference. 2011. https://doi.org/10.1109/EIC.2011.5996173Asiri Ya, Vouk A, Renfort L, Clark D, Copper J. Neural network based classification of partial discharge in hv motors. In: Electrical Insulation Conference. 2011. https://doi.org/10.1109/EIC.2011.5996173Khal’asmaa A.I., Dmitriev S.A., Kokin S.E. An automated system for taking decisions to assess the actual state of electrical equipment // Power Technology and Engineering. 2016. Vol. 49. P. 389–392. https://doi.org/10.1007/s10749-016-0634-6Khal’asmaa AI, Dmitriev SA, Kokin SE. An automated system for taking decisions to assess the actual state of electrical equipment. Power Technology and Engineering. 2016;49:389–392. https://doi.org/10.1007/s10749-016-0634-6Chen Minghua, Liu Qunying, Chen Shuheng, Liu Yicen, Zhang Chang-Hua, Liu Ruihua. XGBoost-based algorithm interpretation and application on post-fault transient stability status prediction of power system // IEEE Access. 2019. Vol. 7. P. 13149–13158. https://doi.org/10.1109/ACCESS.2019.2893448Chen Minghua, Liu Qunying, Chen Shuheng, Liu Yicen, Zhang Chang-Hua, Liu Ruihua. XGBoost-based algorithm interpretation and application on post-fault transient stability status prediction of power system. IEEE Access. 2019;7:13149–13158. https://doi.org/10.1109/ACCESS.2019.2893448Khalyasmaa A.I., Eroshenko S.A. Data analytics platform for power equipment intelligent lifecycle management // Cigre [Электронный ресурс]. URL: http://d2_rus.cigre.ru/reports/0509/D2_315%20Poster_%D0%A4%D0%B0%D0%B9%D0%BB%203.pdf (13.05.2020).Khalyasmaa AI, Eroshenko SA. Data analytics platform for power equipment intelligent lifecycle management. Cigre. Available from: http://d2_rus.cigre.ru/reports/0509/D2_315%20Poster_%D0%A4%D0%B0%D0%B9%D0%BB%203.pdf [Accessed 13th May 2020].Khalyasmaa A.I., Senyuk M.D., Eroshenko S.A. Highvoltage circuit breakers technical state patterns recognition based on machine learning methods // IEEE Transactions on Power Delivery. 2019. Vol. 34. Issue 4 P. 1747– 1756. https://doi.org/10.1109/TPWRD.2019.2921095Khalyasmaa AI, Senyuk MD, Eroshenko SA. Highvoltage circuit breakers technical state patterns recognition based on machine learning methods. IEEE Transactions on Power Delivery. 2019;34(4):1747–1756. https://doi.org/10.1109/TPWRD.2019.2921095Liu Huan, Motoda Hiroshi. Computational methods of feature selection. chapman and Hall/CRC, 2007. 440 p.Liu Huan, Motoda Hiroshi. Computational Methods of Feature Selection. Chapman and Hall/CRC; 2007, 440 p.Khalyasmaa A.I., Senyuk M.D., Eroshenko S.A. Analysis of the state of high-voltage current transformers based on gradient boosting on decision trees // IEEE Transactions on Power Delivery. 2020. Vol. 36. P. 1747– 1756. https://doi.org/10.1109/TPWRD.2020.3021702Khalyasmaa AI, Senyuk MD, Eroshenko SA. Analysis of the state of high-voltage current transformers based on gradient boosting on decision trees. IEEE Transactions on Power Delivery. 2020;36:1747–1756. https://doi.org/10.1109/TPWRD.2020.3021702Salakhutdinova K.I., Lebedev I.S., Krivtsova I.E. Gradient boosting trees method in the task of software identification // Scientific and Technical Journal of Information Technologies, Mechanics and Optics. 2018. Vol. 18. No. 8. P. 1016–1022. https://doi.org/10.17586/2226-1494-2018-18-6-1016-1022Salakhutdinova KI, Lebedev IS, Krivtsova IE. Gradient boosting trees method in the task of software identification. Scientific and Technical Journal of Information Technologies, Mechanics and Optics. 2018;18(8):1016–1022. https://doi.org/10.17586/2226-1494-2018-18-6-1016-1022

The authors declare that there are no conflicts of interest present.