probenerИзвестия высших учебных заведений. ПРОБЛЕМЫ ЭНЕРГЕТИКИPower engineering: research, equipment, technology1998-99032658-5456Kazan State Power Engineering University10.30724/1998-9903-2024-26-1-64-76probener-2965Research ArticleЭЛЕКТРОЭНЕРГЕТИКАELECTRICITYАнсамблевая модель для прогнозирования выработки ветровых электростанцийEnsemble machine learning model for forecasting wind farm generationРусинаА. Г.RusinaA. G.

Анастасия Георгиевна Русина, д-р техн. наук, декан

факультет энергетики

Новосибирск

Anastasia G. Rusina

Novosibirsk

anastasiarusina@gmail.comТувшинОсгонбаатарTuvshinOsgonbaatar

Осгонбаатар Тувшин, аспирант

Новосибирск

Osgonbaatar Tuvshin

Novosibirsk

o.tuvshin.21@gmail.comМатренинП. В.MatreninP. V.

Павел Викторович Матренин, канд. техн. наук, доцент

кафедра систем электроснабжения предприятий

Новосибирск

Pavel V. Matrenin

Novosibirsk

pavel.matrenin@gmail.comСергеевН. Н.SergeevN. N.

Никита Николаевич Сергеев, студент, младший научный сотрудник

Межкафедральная научно-исследовательская лаборатория обработки, анализа и представления данных в электроэнергетических системах

Новосибирск

Nikita N. Sergeev

Novosibirsk

nikita.n.sergeev@gmail.comНовосибирский Государственный Технический УниверситетРоссияNovosibirsk State Technical UniversityRussian Federation2024240420242616476Copyright © Русина А.Г., Тувшин О., Матренин П.В., Сергеев Н.Н., 20242024Русина А.Г., Тувшин О., Матренин П.В., Сергеев Н.Н.Rusina A.G., Tuvshin O., Matrenin P.V., Sergeev N.N.Данная работа распространяется под лицензией Creative Commons Attribution 4.0.This work is licensed under a Creative Commons Attribution 4.0 License.https://www.energyret.ru/jour/article/view/2965

   Применение возобновляемых источников энергии является перспективным путем снижения вредных выбросов в процессе производства энергии и решения проблемы ухудшения экологической ситуации. Несмотря на то, что возобновляемая энергия считается чистой энергией, которую также называют «зеленой» энергией, вопрос ее использования в энергосистеме вызывает определенные трудности. Эффективное использование возобновляемых источников энергии требует актуальной информации о первичных энергоносителях, что особенно важно при крупномасштабной интеграции возобновляемых источников в систему. Таким образом, для обеспечения нормальных режимов работы энергосистемы необходимо прогнозировать выработку возобновляемых источников с допустимой погрешностью.

   ЦЕЛЬ

   ЦЕЛЬ. Прогнозирование суточного графика выработки ветровых электростанций.

   МЕТОДЫ

   МЕТОДЫ. В работе использовались ансамблевые алгоритмы, основанные на деревьях решений и являющиеся одним из подходов машинного обучения. Программная реализация выполнена с помощью языка программирования Python. В качестве исходных данных использованы данные о скорости ветра и выработке определенных электростанций за период 2019-2021 гг.

   РЕЗУЛЬТАТЫ

   РЕЗУЛЬТАТЫ. С помощью предложенной методики был составлен прогноз суточных графиков выработки трех ветровых электростанций с погрешностью от 2,4 до 3,5 МВт или от 5,0 до 7,0 % установленной мощности соответствующих ветровых электростанций. Нормализованная средняя ошибка по модулю в процентах составила от 12,3 до 13,3 %.

   ЗАКЛЮЧЕНИЕ

   ЗАКЛЮЧЕНИЕ. Ансамблевые методы машинного обучения позволяют обнаруживать нелинейные и нестационарные зависимости во временных рядах, а также могут быть реализованы в задаче прогнозирования суточного графика выработки ветроустановок. Повышение точности прогнозирования выработки ветроустановок имеет высокую значимость для эффективного функционирования и планирования режимов энергосистемы.

   The use of renewable energy sources is the only way to avoid emissions from energy production and to decide pollution of ecology. Despite the fact that renewable energy has become clean energy, which called green energy, the issue of using it is quite difficult for the control and regulate the energy system. Efficient use of renewable energy requires information on primary sources. If looking case of large-scale integration, this requirement will be significantly felt. Thus, to ensure the normal operating modes of the energy system, it is necessary to predict the generation of renewable sources with an acceptable error.   PURPOSE. To forecast the generation of wind farms.   METHODS. This study is carried out by ensemble algorithms, such as Random Forest, AdaBoost and XGBoost, which are one of the machine learning approaches. The software implementation is made using the Python programming language. As initial inputs historical data on windspeed and generating of some windfarms in Mongolia by 2019-2021 were used.   RESULTS. The proposed method predicted daily production schedules at three wind farms with an error of 2.4 to 3.4 MW or 5.0 to 7.0 percent of the installed capacity of the corresponding wind farm. Also, normalized MAE was 12,3 to 13.3 percent.   CONCLUSIONS. Ensemble methods of machine learning made it possible to determine non-linear and non-stationary dependencies of the time series, and also can be implemented in the problem of predicting the daily production schedule. Increasing the accuracy of wind energy forecasting will affect positively the operation and planning of the power systems.

прогнозированиевыработкаскорость ветраансамблевый методветровая электростанцияforecastinggeneratewindspeedensemble methodwind farmReferencesOlaofe Z. O. A 5-day wind speed and power forecasts using a layer recurrent neural network (LRNN) // Sustainable Energy Technologies and Assessments. 2014. Vol. 6. P. 1-24. doi: 10.1016/j.seta.2013.12.001Olaofe Z. O. A 5-day wind speed and power forecasts using a layer recurrent neural network (LRNN). Sustainable Energy Technologies and Assessments, 2014. vol. 6, pp.1-24. doi: 10.1016/j.seta.2013.12.001Santos M., González M. Factors that influence the performance of wind farms // Renewable Energy. 2019. Vol. 135. P. 643-651. doi: 10.1016/j.renene.2018.12.033Santos M., González M. Factors that influence the performance of wind farms, Renewable Energy, 2019, vol.135, pp. 643-651. doi: 10.1016/j.renene.2018.12.033Ouyang T., Kusiak A., He Y. Predictive model of yaw error in a wind turbine // Energy. 2017. Vol. 123. P. 119-130. doi: 10.1016/j.energy.2017.01.150Ouyang T., Kusiak A., He Y. Predictive model of yaw error in a wind turbine. Energy. 2017. vol. 123. pp. 119-130. doi: 10.1016/j.energy.2017.01.150Matrenin P. V., Osgonbaatar T., Sergeev N. N. Overview of Renewable Energy Sources in Mongolia // 2022 IEEE International Multi-Conference on Engineering, Computer and Information Sciences (SIBIRCON). IEEE. 2022. P. 700-703. doi: 10.1109/SIBIRCON56155.2022.10016986Matrenin P. V., Osgonbaatar T., Sergeev N. N. Overview of Renewable Energy Sources in Mongolia. 2022 IEEE International Multi-Conference on Engineering, Computer and Information Sciences (SIBIRCON). IEEE, 2022. pp. 700-703. doi: 10.1109/SIBIRCON56155.2022.10016986Zhao L, Muhammad S.N., Hafiz M.J., Ahmed N.A. A review on proliferation of artificial intelligence in wind energy forecasting and instrumentation management // Environmental Science and Pollution Research. 2022. №. 29. Vol. 29. P. 43690-43709. doi: 10.1007/s11356-022-19902-8Zhao L, Muhammad S.N., Hafiz M.J., Ahmed N.A. A review on proliferation of artificial intelligence in wind energy forecasting and instrumentation management. Environmental Science and Pollution Research. 2022. vol. 29. pp. 43690-43709. doi: 10.1007/s11356-022-19902-8Manusov V., Matrenin P., Nazarov M., Beryozkina S. Short-Term Prediction of the Wind Speed Based on a Learning Process Control Algorithm in Isolated Power Systems // Sustainability. 2023. №. 2. Vol. 15. P. 1730. doi: 10.3390/su15021730Manusov V., Matrenin P., Nazarov M., Beryozkina S. Short-Term Prediction of the Wind Speed Based on a Learning Process Control Algorithm in Isolated Power Systems. Sustainability. 2023. vol. 2(15). pp. 1730. doi: 10.3390/su15021730Tiwari S. Wind speed forecasting methods for wind energy generation // 2022 1<sup>st</sup> International Conference on Informatics (ICI). IEEE, 2022. P. 143-147. doi: 10.1109/ICI53355.2022.9786880Tiwari, Seemant. Wind speed forecasting methods for wind energy generation. 2022 1<sup>st</sup> International Conference on Informatics (ICI). IEEE. 2022. pp. 143-147. doi: 10.1109/ICI53355.2022.9786880Weather forecasting for renewable energy system: a review / Meenal R. et al. // Archives of Computational Methods in Engineering. 2022. №. 5. Vol. 29. P. 2875-2891. doi: 10.1007/s11831-021-09695-3Meenal, R., et al. Weather forecasting for renewable energy system: a review. Archives of Computational Methods in Engineering. 2022. vol. 5(29). pp. 2875-2891. doi: 10.1007/s11831-021-09695-3Li X., Sabas J. F., Mendéz V. D. Wind energy forecasting using multiple ARIMA models // 2022 IEEE 18<sup>th</sup> International Conference on Automation Science and Engineering (CASE). IEEE, 2022. P. 2034-2039. doi: 10.1109/CASE49997.2022.9926516Li, Xiaoou, Juan Francisco Sabas, and Vicente Duarte Mendéz. Wind energy forecasting using multiple ARIMA models. 2022 IEEE 18<sup>th</sup> International Conference on Automation Science and Engineering (CASE). IEEE, 2022. pp. 2034-2039. doi: 10.1109/CASE49997.2022.9926516A survey of artificial neural network in wind energy systems / Marugán A. P. et al // Applied energy. 2018. Vol. 228. P. 1822-1836. doi: 10.1016/j.apenergy.2018.07.084Marugán A. P. et al. A survey of artificial neural network in wind energy systems. Applied energy 228. 2018. pp.1822-1836. Doi: 10.1016/j.apenergy.2018.07.084Zhang Y., Li R. Short term wind energy prediction model based on data decomposition and optimized LSSVM // Sustainable Energy Technologies and Assessments. 2022. Vol. 52. P. 102025. doi: 10.1016/j.seta.2022.102025Zhang Y., and Li R. Short term wind energy prediction model based on data decomposition and optimized LSSVM. Sustainable Energy Technologies and Assessments. 2022. vol.52. pp. 102025. doi: 10.1016/j.seta.2022.102025Regression model for predicting the speed of wind flows for energy needs based on fuzzy logic /Khasanzoda N. et al // Renewable Energy. 2022. Vol. 191. P. 723-731. doi: 10.1016/j.renene.2022.04.017Khasanzoda, Nasrullo, et al. Regression model for predicting the speed of wind flows for energy needs based on fuzzy logic. Renewable Energy. 2022. vol. 191. pp. 723-731. doi: 10.1016/j.renene.2022.04.017A new wind power prediction method based on chaotic theory and Bernstein Neural Network / Wang C. и др.// Energy. 2016. Vol. 117. P. 259-271. doi: 10.1016/j.energy.2016.10.041Wang C, et.al. A new wind power prediction method based on chaotic theory and Bernstein Neural Network. Energy. 2016. vol. 117. pp. 259-271. doi: 10.1016/j.energy.2016.10.041Buhan S., Özkazanç Y., Çadırcı I. Wind pattern recognition and reference wind mast data correlations with NWP for improved wind-electric power forecasts // IEEE Transactions on Industrial Informatics. 2016. №. 3. Vol. 12. P. 991-1004. doi: 10.1109/TII.2016.2543004Buhan S., Özkazanç Y., Çadırcı I. Wind pattern recognition and reference wind mast data correlations with NWP for improved wind-electric power forecasts. IEEE Transactions on Industrial Informatics. 2016. vol. 3(12). pp. 991-1004. doi: 10.1109/TII.2016.2543004Negative correlation learning-based RELM ensemble model integrated with OVMD for multi-step ahead wind speed forecasting / Peng T. et al // Renewable Energy. 2020. Vol. 156. P. 804-819. doi: 10.1016/j.renene.2020.03.168Peng T., et.al. Negative correlation learning-based RELM ensemble model integrated with OVMD for multi-step ahead wind speed forecasting. Renewable Energy. 2020. vol.156. pp. 804-819. doi: 10.1016/j.renene.2020.03.168Smart wind speed forecasting approach using various boosting algorithms, big multi-step forecasting strategy / Li Y. et al // Renewable energy. 2019. Мўд. 135. З. 540-553. doi: 10.1016/j.renene.2018.12.035Li Y., et.al. Smart wind speed forecasting approach using various boosting algorithms, big multi-step forecasting strategy. Renewable energy. 2019. vol.135. pp.540-553. doi: 10.1016/j.renene.2018.12.035Wang L., Guo Y., Li X. Wind speed prediction using measurements from neighboring locations and combining the extreme learning machine and the AdaBoost algorithm // Energy Reports. 2022. Vol. 8. P. 1508-1518. doi: 10.1016/j.egyr.2021.12.062Wang L., Guo Y., Li X. Wind speed prediction using measurements from neighboring locations and combining the extreme learning machine and the AdaBoost algorithm. Energy Reports. 2022. vol. 8. pp. 1508-1518. doi: 10.1016/j.egyr.2021.12.062Dosdoğru A. T., İpek A. B. Hybrid boosting algorithms and artificial neural network for wind speed prediction // International Journal of Hydrogen Energy. 2022. №. 3. Vol. 47. P. 1449-1460. doi: 10.1016/j.ijepes.2021.107365Dosdoğru A. T., İpek A. B. Hybrid boosting algorithms and artificial neural network for wind speed prediction. International Journal of Hydrogen Energy. 2022. no. 3. vol. 47. pp. 1449-1460. doi: 10.1016/j.ijepes.2021.107365A multi-level model for hybrid short term wind forecasting based on SVM, wavelet transform and feature selection / Abedinia O. et al // 2022 IEEE International Conference on Environment and Electrical Engineering and 2022 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe). IEEE, 2022. P. 1-6. doi: 10.1109/EEEIC/ICPSEurope54979.2022.9854519Abedinia O., et.al. A multi-level model for hybrid short term wind forecasting based on SVM, wavelet transform and feature selection. 2022 IEEE International Conference on Environment and Electrical Engineering and 2022 IEEE Industrial and Commercial Power Systems Europe (EEEIC/I&CPS Europe). IEEE, 2022. pp. 1-6. doi: 10.1109/EEEIC/ICPSEurope54979.2022.9854519Lahouar A., Slama J. B. H. Hour-ahead wind power forecast based on random forests // Renewable energy. 2017. Vol. 109. P. 529-541. doi: 10.1016/j.renene.2017.03.064Lahouar A., Slama J. B. H. Hour-ahead wind power forecast based on random forests. Renewable energy. 2017. vol. 109. pp. 529-541. doi: 10.1016/j.renene.2017.03.064Breiman L. Random forests // Machine learning. 2001. Vol. 45. P. 5-32. doi: 10.1023/a:1010933404324Breiman, Leo. Random forests. Machine learning. 2001. vol. 45. pp. 5-32. doi: 10.1023/a:1010933404324Jorgensen K. L., Shaker H. R. Wind power forecasting using machine learning: State of the art, trends and challenges // 2020 IEEE 8th International Conference on Smart Energy Grid Engineering (SEGE). IEEE, 2020. P. 44-50. doi: 10.1109/SEGE49949.2020.9181870Jorgensen K. L., Shaker H. R. Wind power forecasting using machine learning: State of the art, trends and challenges. 2020 IEEE 8th International Conference on Smart Energy Grid Engineering (SEGE). IEEE, 2020. pp. 44-50. doi: 10.1109/SEGE49949.2020.9181870Freund Y., Schapire R. E. A decision-theoretic generalization of on-line learning and an application to boosting // Journal of computer and system sciences. 1997. №. 1. Vol. 55. P. 119-139. doi: 10.1006/jcss.1997.1504Freund Y., Schapire R. E. A decision-theoretic generalization of on-line learning and an application to boosting. Journal of computer and system sciences. 1997. no. 1. vol. 55. pp. 119-139. doi: 10.1006/jcss.1997.1504Phan Q. T., Wu Y. K., Phan Q. D. A hybrid wind power forecasting model with XGBoost, data preprocessing considering different NWPs // Applied Sciences. 2021. №. 3. Vol. 11. P. 1100. doi: 10.3390/app11031100Phan Q. T., Wu Y. K., Phan Q. D. A hybrid wind power forecasting model with XGBoost, data preprocessing considering different NWPs. Applied Sciences. 2021. no. 3. vol. 11. pp. 1100. doi: 10.3390/app11031100Ouyang T., Kusiak A., He Y. Modeling wind-turbine power curve: A data partitioning and mining approach // Renewable Energy. 2017. Vol. 102. P. 1-8. doi: 10.1016/j.renene.2016.10.032Ouyang T., Kusiak A., He Y. Modeling wind-turbine power curve: A data partitioning and mining approach. Renewable Energy. 2017. vol. 102. pp. 1-8. doi: 10.1016/j.renene.2016.10.032

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