1. На главную
  2. Электронные версии
  3. Доклады Российской академии наук. Математика, информатика, процессы управления
  4. T. 514, № 2, 2023

Доклады Российской академии наук. Математика, информатика, процессы управления, 2023, T. 514, № 2, стр. 318-332

ОБЪЕДИНЯЯ ПРОГНОСТИЧЕСКОЕ ПЛАНИРОВАНИЕ И ОБЛАЧНЫЕ ВЫЧИСЛЕНИЯ ДЛЯ СНИЖЕНИЯ ВЫБРОСОВ УГЛЕКИСЛОГО ГАЗА ПРИ ОБУЧЕНИИ МОДЕЛЕЙ МАШИННОГО ОБУЧЕНИЯ

М. Тютюльников 1, В. Лазарев 1, А. Коровин 1, Н. Захаренко 2, И. Дорощенко 2, С. Буденный 1, 2, *

1 Научно-исследовательский институт искусственного интеллекта (AIRI)
Москва, Россия

2 Сбер
Москва, Россия

* E-mail: sanbudenny@sberbank.ru

Поступила в редакцию 02.09.2023
После доработки 15.09.2023
Принята к публикации 18.10.2023

Аннотация

Мы представляем eco4cast1, пакет с открытым исходным кодом, предназначенный для снижения углеродного следа моделей машинного обучения с помощью прогностического планирования облачных вычислений. Пакет интегрируется в модели машинного обучения и использует разработанную временную сверточную нейронную сеть (TCN) для прогнозирования суточной углеродоемкости электроэнергии. Высокая точность прогнозирования модели достигается за счет учета погодных условий, обладающих устойчивой корреляцией с углеродоемкостью. Задачей eco4cast является способность определять временные периоды минимальной углеродоемкости электроэнергии. Это позволяет пакету назначать задачи облачных вычислений только на эти периоды, снижая воздействие моделей на окружающую среду. Роль пакета в уменьшении эмиссии состоит в сочетании экологичности вычислений и их вычислительной эффективности. Код и документация пакета размещены на Github под лицензией Apache 2.0.

Ключевые слова: ESG, устойчивый ИИ, зеленый ИИ, устойчивое развитие, экология, углеродный след, эмиссия CO2, рациональное планирование

Список литературы

  1. Paris Agreement. Paris agreement. In Report of the Conference of the Parties to the United Nations Framework Convention on Climate Change (21st Session, 2015: Paris). Retrived December, volume 4, page 2017. HeinOnline, 2015.

  2. Xiaoyuan Wang, Jiahaoran Wang, Weimin Guan, Farhad Taghizadeh-Hesary. Role of esg investments in achieving cop-26 targets. Energy Economics. 2023. V. 123. P. 106757.

  3. Gibon T., Hertwich E.G., Arvesen A., Singh B., Verones F. Health benefits, ecological threats of low-carbon electricity. Environmental Research Letters. 2017. V. 12 (3). P. 034023.

  4. Pesce M. Cloud computing’s coming energy crisis. IEEE Spectrum, 2021.

  5. Henderson P., Hu J., Romoff J., Brunskill E., Jurafsky D., Pineau J. Towards the systematic reporting of the energy and carbon footprints of machine learning. Journal of Machine Learning Research. 2020. V. 21 (248). P. 1–43.

  6. Patterson D., Gonzalez J., Le Q., Liang C., Munguia L.-M., Rothchild D., So D., Texier M., Dean J. Carbon emissions and large neural network training. arXiv preprint arXiv:2104.10350, 2021.

  7. Yanchao Feng, Juan Zhang, Yong Geng, Shurui Jin, Ziyi Zhu, Zhou Liang. Explaining and modeling the reduction effect of low-carbon energy transition on energy intensity: Empirical evidence from global data. Energy. 2023. V. 281. P. 128276.

  8. Budennyy S.A., Lazarev V.D., Zakharenko N.N., Korovin A.N., Plosskaya O.A., Dimitrov D.V., Akhripkin V.S., Pavlov I.V., Oseledets I.V., Barsola I.S., Egorov I.V., Kosterina A.A., Zhukov L.E. eco2ai: Carbon emissions tracking of machine learning models as the first step towards sustainable ai. Doklady Mathematics, 2023.

  9. Mohammad Mahdi Forootan, Iman Larki, Rahim Zahedi, Abolfazl Ahmadi. Machine learning and deep learning in energy systems: A review. Sustainability. 2022. V. 14 (8).

  10. Zhou Xuan, Zi Xuehui, Liang Liequan, Fan Zubing, Yan Junwei, Pan Dongmei. Forecasting performance comparison of two hybrid machine learning models for cooling load of a large-scale commercial building. Journal of Building Engineering. 2019. V. 21. P. 64–73.

  11. Runge J., Radu Zmeureanu, Mathieu Le Cam. Hybrid short-term forecasting of the electric demand of supply fans using machine learning. Journal of Building Engineering. 2020. V. 29. P. 101144.

  12. Hussein Sharadga, Shima Hajimirza, Balog R.S. Time series forecasting of solar power generation for large-scale photovoltaic plants. Renewable Energy. 2020. V. 150. P. 797–807.

  13. Cunbin Li, Shuaishuai Lin, Fangqiu Xu, Ding Liu, Jicheng Liu. Short-term wind power prediction based on data mining technology and improved support vector machine method: A case study in northwest china. Journal of Cleaner Production. 2018. V. 205. P. 909–922.

  14. Wendong Yang, Jianzhou Wang, Haiyan Lu, Tong Niu, Pei Du. Hybrid wind energy forecasting and analysis system based on divide and conquer scheme: A case study in china. Journal of Cleaner Production. 2019. V. 222. P. 942–959.

  15. Dash P.K., Eluri N.V., Prasad D.V., Ravi Kumar Jalli, Mishra S.P. Multiple power quality disturbances analysis in photovoltaic integrated direct current microgrid using adaptive morphological filter with deep learning algorithm. Applied Energy. 2022. V. 309. P. 118454, 2022.

  16. Elissaios Sarmas, Evangelos Spiliotis, Vangelis Marinakis, Themistoklis Koutselis, Haris Doukas. A meta-learning classification model for supporting decisions on energy efficiency investments. Energy and Buildings. 2022. V. 258. P. 111836.

  17. L’eonard Tschora, Erwan Pierre, Marc Plantevit, C’eli-ne Robardet. Electricity price forecasting on the day-ahead market using machine learning. Applied Energy. 2022. V. 313. P. 118752.

  18. Hsiao-Tien Pao, Hsin-Chia Fu, Cheng-Lung Tseng. Forecasting of co2 emissions, energy consumption and economic growth in china using an improved grey model. Energy. 2012. V. 40 (1). P. 400–409.

  19. Surbhi Kumari, Sunil Kumar Singh. Machine learning-based time series models for effective co2 emission prediction in india. Environmental Science and Pollution Research, 2022.

  20. Yang Meng, Hossain Noman. Predicting co2 emission footprint using ai through machine learning. Atmosphere. 2022. V. 13 (11).

  21. Zhili Zuo, Haixiang Guo, Jinhua Cheng. An lstm-stripat model analysis of china’s 2030 co2 emissions peak. Carbon Management. 2020. V. 11 (6). P. 577–592, 2020.

  22. Bonga Wellington Garikai, Thabani Nyoni. Prediction of co2 emissions in india using arima models. DRJ – Journal of Economics & Finance. 2019. V. 4. P. 01–10.

  23. Pooja Gopu, Rama Ranjan Panda, Naresh Kumar Nagwani. Time series analysis using arima model for air pollution prediction in hyderabad city of india. In V. Sivakumar Reddy, V. Kamakshi Prasad, Jiacun Wang, K. T. V. Reddy, editors, Soft Computing and Signal Processing, pages 47–56, Singapore. Springer Singapore, 2021.

  24. Huiru Zhao, Guo Huang, Ning Yan. Forecasting energy-related co2 emissions employing a novel ssa-lssvm model: considering structural factors in china. Energies. 2018. V. 11 (4). P. 781.

  25. Mohammed Redha Qader, Shahnawaz Khan, Musta-fa Kamal, Muhammad Usman, Mohammad Haseeb. Forecasting carbon emissions due to electricity power generation in bahrain. Environmental Science and Pollution Research. 2022. V. 29 (12). P. 17346–17357.

  26. Melahat Sevgül Bakay, Ümit Ağbulut. Electricity production based forecasting of greenhouse gas emissions in turkey with deep learning, support vector machine and artificial neural network algorithms. Journal of Cleaner Production. 2021. V. 285. P. 125324.

  27. Chairul Saleh, Nur Rachman Dzakiyullah, Jonathan Bayu Nugroho. Carbon dioxide emission prediction using support vector machine. IOP Conference Series: Materials Science and Engineering. 2016. V. 114 (1). P. 012148.

  28. Kenneth Leerbeck, Peder Bacher, Rune Gronborg Junker, Goran Goranovi’c, Olivier Corradi, Razgar Ebrahimy, Anna Tveit, Henrik Madsen. Short-term forecasting of co2 emission intensity in power grids by machine learning. Applied Energy. 2020. V. 277. P. 115527.

  29. David Patterson, Joseph Gonzalez, Urs Hölzle, Quoc Le, Chen Liang, Lluis-Miquel Munguia, Daniel Rothchild, David So, Maud Texier, Jeff Dean. The carbon footprint of machine learning training will plateau, then shrink, 2022.

  30. Jin-Young Kim, Sung-Bae Cho. Electric energy consumption prediction by deep learning with state explainable autoencoder. Energies. 2019. V. 12 (4).

  31. Ernesto Aguilar Madrid, Nuno Antonio. Short-term electricity load forecasting with machine learning. Information. 2021. V. 12 (2).

  32. Tahseen Khan, Wenhong Tian, Shashikant Ilager, Rajkumar Buyya. Workload forecasting and energy state estimation in cloud data centres: Ml-centric approach. Future Generation Computer Systems. 2022. V. 128. P. 320–332.

  33. Ruilong Deng, Zaiyue Yang, Jiming Chen, Navid Rahbari Asr, and Mo-Yuen Chow. Residential energy consumption scheduling: A coupled-constraint game approach. IEEE Transactions on Smart Grid. 2014. V. 5 (3). P. 1340–1350.

  34. Yuan Hong, Shengbin Wang, and Ziyue Huang. Efficient energy consumption scheduling: Towards effective load leveling. Energies. 2017. V. 10 (1).

  35. Karin van der Wiel, Hannah C. Bloomfield, Robert W. Lee, Laurens P. Stoop, Russell Blackport, James A. Screen, and Frank M. Selten. The influence of weather regimes on european renewable energy production and demand. Environmental Research Letters. 2019. V. 14 (9). P. 094010.

  36. Patrick Zippenfenig. Open-meteo.com weather api. July 2023.https://doi.org/10.5281/zenodo.8112599

  37. Met Office. Cartopy: a cartographic python library with a Matplotlib interface. Exeter, Devon, 2010–2015.

  38. Shaojie Bai, J. Zico Kolter, Vladlen Koltun. An empi-rical evaluation of generic convolutional and recur-rent networks for sequence modeling. arXiv preprint arXiv:1803.01271, 2018.

Дополнительные материалы отсутствуют.

Инструменты

следующая статья выпуска предыдущая статья выпуска содержание выпуска

Доклады Российской академии наук. Математика, информатика, процессы управления

Архивы выпусков Информация о журнале Отправить рукопись в журнал