APPLICATION OF MACHINE LEARNING METHODS FOR TIME SERIES FORECASTING IN ENERGY RESOURCE CONSUMPTION PLANNING
DOI:
https://doi.org/10.52754/16948645_2023_1_220Keywords:
forecasting, multivariate time series, energy consumption planning, machine learning, deep learning, interval forecasting.Abstract
The article deals with the actual problem of forecasting the consumption of energy resources, in particular, the volume of thermal energy consumption for residential buildings. The main aspects of choosing a forecasting model depending on the formulation of the problem and the nature of the forecasted data are presented. Several modern methods of predicting multidimensional time series are considered and the accuracy of forecast models based on real practical data is studied, and the precision is compared with models based on classical statistical forecasting methods.
References
Manusov V. Analysis of electricity consumption forecasting methods for the coal industry / V. Manusov, D. Orlov, V. Karmanov [et al.] // Przeglad Elektrotechniczny. – 2022. – Vol. 98, iss. 9. – P. 26-31. DOI 10.15199/48.2022.09.05 DOI: https://doi.org/10.15199/48.2022.09.05
Manusov V. Forecasting Electricity Consumption of Electrical Machines of a Coal Industry Enterprise Using the Wavelet Transform / V. Manusov, D. Orlov, V. Karmanov [et. al.]. // IEEE 23 International Conference of Young Professionals in Electron Devices and Materials (EDM) to the 100th anniversary of the legendary NETI rector Georgy Lyshchinsky : IEEE, 2022. – 4 p. – ISBN 978-1-6654-9804-3. DOI 10.1109/EDM55285.2022.9855175 DOI: https://doi.org/10.1109/EDM55285.2022.9855175
Manusov V. Investigation of Load Schedules of Electrical Machines of a Mining Enterprise Using Wavelet Analysis / V. Z. Manusov, D. V. Orlov, P. V. Matrenin [et al.] // IEEE 23 International Conference of Young Professionals in Electron Devices and Materials (EDM) to the 100th anniversary of the legendary NETI rector Georgy Lyshchinsky : IEEE, 2022. – 4 p. – ISBN 978-1-6654-9804-3. DOI: 10.1109/EDM55285.2022.9855045 DOI: https://doi.org/10.1109/EDM55285.2022.9855045
Manusov V. Predictive Control and Production Process Forecasting Under Deterministic Chaos / V. Manusov, D. Orlov, V. Karmanov [et. al.] // IEEE 23 International Conference of Young Professionals in Electron Devices and Materials (EDM) to the 100th anniversary of the legendary NETI rector Georgy Lyshchinsky : IEEE, 2022. – 5 p. – ISBN 978-1-6654-9804-3. DOI: 10.1109/EDM55285.2022.9855192 DOI: https://doi.org/10.1109/EDM55285.2022.9855192
Hastie T. The Elements of Statistical Learning / T. Hastie, R. Tibshirani, J. Friedman. – Springer, 2017. – 763 p.
Brownlee J. Introduction to Time Series Forecasting with Python / J. Brownlee. – Machine learning mastery, 2020. – 364 p.
Manusov V. Analysis of methods of electricity consumption forecasting for a coal industry enterprise / V. Manusov, D. Orlov, V. Karmanov [et al.] // IOP Conference Series: Earth and Environmental Science. – 2022. – Vol. 1045, iss. 1 : Ensuring sustainable development in the context of agriculture, green energy, ecology and earth science (ESDCA–2022): 2 International scientific and practical conference, Smolensk, 23–27 Jan. 2022. – Art. 012035 (11 p.). DOI 10.1088/1755-1315/1045/1/012035. DOI: https://doi.org/10.1088/1755-1315/1045/1/012035
Айвазян С. А. Прикладная статистика. Основы эконометрики : учебник для вузов : в 2-х т. / С. А. Айвазян. – М.: ЮНИТИ-ДАНА, 2001. – Т. 2. – 432 с.
Rusyana A. Application of Clustering and VARIMA for Rainfall Prediction / A. Rusyana, N. Tatsara // IOP Conference Series Materials Science and Engineering. 2020. 796(1):012063. DOI:10.1088/1757-899X/796/1/012063 DOI: https://doi.org/10.1088/1757-899X/796/1/012063
Natekin A. Gradient boosting machines, a tutorial / A. Natekin, A. Knoll // Front. Neurorobot. 2013. Vol. 7. № 21. DOI: 10.3389/fnbot.2013.00021 DOI: https://doi.org/10.3389/fnbot.2013.00021
Anghel A. Benchmarking and optimization of gradient boosting decision tree algorithms / A. Anghel, N. Papandreou, T. Parnell [et al.] – URL: https://arxiv.org/abs/1809.04559 (дата обращения – 30.03.2023 г.)
Elman J. Finding Structure in Time / J. Elman // Cognitive Science. 1990. 179-211 pp. DOI: https://doi.org/10.1207/s15516709cog1402_1
Hochreiter S. Long Short-Term Memory / S. Hochreiter, J. Schumidhuber // Neural Computation. 1997. Vol. 9. № 8. 1735-1780 pp. DOI: 10.1162/neco.1997.9.8.1735 DOI: https://doi.org/10.1162/neco.1997.9.8.1735
Vaswani A. Attention Is All You Need / A. Vaswani, N. Shazeer, N. Parmar [et al.] – URL: https://arxiv.org/abs/1706.03762 (дата обращения – 30.03.2023 г.)
Lim B. Temporal Fusion Transformers for Interpretable Multi-horizon Time Series Forecasting / B. Lim, S. Arik, N. Loeff [et al.] – URL: https://arxiv.org/abs/1912.09363 (дата обращения – 30.03.2023 г.)
Bai S. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling / S. Bai, J. Kolter, V. Koltun – URL: https://arxiv.org/abs/1803.01271 (дата обращения – 30.03.2023 г.)
Сапрыкин К. Автоматизация процесса анализа временных рядов с использованием модели AUTO.ARIMA в R / К. Сапрыкин // Научный журнал. 2019. №5 (39). URL: https://cyberleninka.ru/article/n/avtomatizatsiya-protsessa-analiza-vremennyh-ryadov-s-ispolzovaniem-modeli-auto-arima-v-r (дата обращения: 30.03.2023 г.)
Zorin P. Data of heating meters from residential buildings in Tomsk (Russia) for statistical modeling of the thermal characteristics of buildings / P. Zorin, O. Stukach – URL: https://ieee-dataport.org/documents/data-heating-meters-residential-buildings-tomsk-russia-statistical-modeling-thermal (дата обращения – 30.03.2023 г.)
Biloš M. Irregularly-Sampled Time Series Modeling with Spline Networks / M. Biloš, E. Ramneantu, S. Günnemann– URL: https://arxiv.org/abs/2210.10630 (дата обращения – 30.03.2023 г.)
CatBoost [Electronic resource] / URL: https://catboost.ai/docs/ (date of the application: 01.03.2023)
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2023 Journal of Osh State University. Mathematics. Physics. Technical Sciences
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
