Increasing the accuracy of forecasting the electricity consumption of an industrial enterprise by machine learning methods using the selection of significant features from a time series

This study aims to improve the accuracy of forecasting the electricity consumption of an enterprise based on an analysis and preliminary processing of input data, as well as at evaluating the effect caused by feature selection on the results of various forecast models. A woodworking enterprise located in Nizhniy Novgorod was selected as a forecast object. Two types of machine learning methods, including neural network and ensemble models, were compared. An approach to selecting the most significant parameters (features) from a time series was considered in order to improve the results of the following ensemble models based on decision trees: adaptive busting (AdaBoost), Gradient Boosting and Random Forest. The most significant features of the initial time series were determined using the calculation of correlation coefficients between the values of electricity consumption in forecasted and previous hours. For the considered forecast object, the most significant features were established to be the consumed energy in hours lagging behind the forecasted hour by the multiple number of days. The schedule of repair works for woodworking machines was used as an additional feature. According to the obtained results, decision tree ensembles can surpass artificial neural networks provided that significant features are selected correctly. Thus, the smallest average error of a neural network model on a test sample comprised 7.0%, while an error of 5.5% was obtained for a Gradient Boosting ensemble model. The use of a repair schedule was demonstrated to additionally increase the forecast accuracy: for the considered ensemble models, the error reduced from 20 to 30%.

1. Lozinskaia A., Redkina A., Shenkman E. Electricity consumption forecasting for integrated power system with seasonal patterns. Applied Econometrics. 2020;60:5-25. https://doi.org/10.22394/1993-7601-2020-60-5-25.

2. Filippova T. A., Rusina A. G., Dronova Yu. V. Models and methods of electric energy and power forecasting in power system operation management: monograph. Novo-sibirsk: Novosibirsk State Technical University; 2009, 368 p. (In Russ.).

3. Hernandez L., Baladron C., Aguiar J. M., Carro B., Sanchez-Esguevillas A. J., Lloret J., et al. A survey on electric power demand forecasting: future trends in smart grids, microgrids and smart buildings. IEEE Communica-tions Surveys & Tutorials. 2014;16(3):1460-1495. https://doi.org/10.1109/SURV.2014.032014.00094.

4. Deb Chirag, Zhang Fan, Yang Junjing, Lee Siew Eang, Shah Kwok Wei. A review on time series forecasting tech-niques for building energy consumption. Renewable and Sustainable Energy Reviews. 2017;74:902-924. https://doi.org/10.1016/j.rser.2017.02.085.

5. Hahn Heiko, Meyer-Nieberg Silja, Pickl Stefan. Electric load forecasting methods: tools for decision making. Eu-ropean Journal of Operational Research. 2009;199(3):902-907. https://doi.org/10.1016/j.ejor.2009.01.062.

6. Cai M., Pipattanasomporn M., Rahman S. Day-ahead building-level load forecasts using deep learning vs tradi-tional time-series techniques. Applied Energy. 2019;236:1078-1088. https://doi.org/10.1016/j.apenergy.2018.12.042.

7. Qiu Xueheng, Zhang Le, Ren Ye, Suganthan P. N., Amaratunga G. Ensemble deep learning for regression and time series forecasting. In: IEEE Symposium on Computational Intelligence in Ensemble Learning. 2014. https://doi.org/10.1109/CIEL.2014.7015739.

8. Polyakhov N. D., Prikhodko I. A., Efen Van. Electric load forecasting based on support vector machine opti-mized. Izvestiya Sankt-Peterburgskogo gosudarstvennogo elektrotekhnicheskogo universiteta. 2014;10:26-30. (In Russ.).

9. Yildiz B., Bilbao J. I., Sproul A. B. A review and analysis of regression and machine learning models on commer-cial building electricity load forecasting. Renewable and Sustainable Energy Reviews. 2017;73:1104-1122. https://doi.org/10.1016/j.rser.2017.02.023.

10. Yang Jingfei, Stenzel J. Short-term load forecasting with increment regression tree. Electric Power Systems Research. 2006;76(9-10):880-888. https://doi.org/10.1016/j.epsr.2005.11.007.

11. Matrenin P., Safaraliev M., Dmitriev S., Kokin S., Ghu-lomzoda A., Mitrofanov S. Medium-term load forecasting in isolated power systems based on ensemble machine learning models. Energy Reports. 2022;8:612-618. https://doi.org/10.1016/j.egyr.2021.11.175.

12. Matrenin P., Antonenkov D., Manusov V. Recurrent and ensemble models for short-term load forecasting of coal mining companies. In: Ural-Siberian Smart Energy Conference. 2021. https://doi.org/10.1109/USSEC53120.2021.9655732.

13. Neupane B., Perera K. S., Aung Zeyar, Woon Wei Lee. Artificial neural network-based electricity price forecasting for smart grid deployment. In: IEEE International Conference on Com-puter Systems and Industrial Informatics. 2012. https://doi.org/10.1109/ICCSII.2012.6454392.

14. Amarasinghe K., Marino D. L., Manic M. Deep neural networks for energy load forecasting. In: IEEE 26th Inter-national Symposium on Industrial Electronics. 19–21 June 2017, Edinburgh. Edinburgh: IEEE; 2017, p. 1483-1488. https://doi.org/10.1109/ISIE.2017.8001465.

15. Jie Cai, Jiawei Luo, Shulin Wang, Sheng Yang. Feature selection in machine learning: a new perspective. Neurocomputing. 2018;300:70-79. https://doi.org/10.1016/j.neucom.2017.11.077.

16. Koprinska I., Rana M., Agelidis V. G. Correlation and instance based feature selection for electricity load fore-casting. Knowledge-Based Systems. 2015;82:29-40. https://doi.org/10.1016/j.knosys.2015.02.017.

17. Bouktif S., Fiaz A., Ouni A., Serhani M. A. Optimal deep learning LSTM model for electric load forecasting using feature selection and genetic algorithm: comparison with machine learning approaches. Energies. 2018;11(7):1636. https://doi.org/10.3390/en11071636.

18. Huber P. J. Robust estimation of a location parameter. The Annals of Mathematical Statistics. 1964;35(1):73-101. https://doi.org/10.1214/aoms/1177703732.

19. Pedregosa F., Varoquaux G., Gramfort A., Michel V., Thirion B., Grisel O., et al. Scikit-learn: machine learning in Python. Journal of Machine Learning Research. 2011;12(85):2825-2830.

20. Ramachandran P., Zoph B., Le Quoc V. Searching for activation functions. 2017. https://doi.org/10.48550/arXiv.1710.05941.

21. Hoffer E., Hubara I., Soudry D. Train longer, general-ize better: closing the generalization gap in large batch training of neural networks. 2017. https://doi.org/10.48550/arXiv.1705.08741.

22. Khirirat S., Feyzmahdavian H. R., Johansson M. Mini-batch gradient descent: faster convergence under data sparsity. In: IEEE 56th Annual Conference on Decision and Control. 12–15 December 2017, Melbourne. Mel-bourne: IEEE; 2017, p. 2880-2887. https://doi.org/10.1109/CDC.2017.8264077.

23. Linjordet T., Balog K. Impact of training dataset size on neural answer selection models. In: Azzopardi L., Stein B., Fuhr N., Mayr P., Hauff C., Hiemstra D. (eds.). Ad-vances in Information Retrieval. Lecture Notes in Com-puter Science. Vol. 11437. Cham: Springer; 2019, р. 828-835. https://doi.org/10.1007/978-3-030-15712-8_59.