Abstract
Intelligent energy saving and energy efficient technologies are a modern large-scale global trend in the development of energy systems. Accurate estimates of energy savings are important for promoting energy-efficient construction projects and demonstrating their economic potential. The growing digital measurement infrastructure used in commercial buildings increases the availability of high-frequency data that can be employed for anomaly detection, diagnostics of equipment, heating systems, and ventilation, as well as optimization of air conditioning. This implies the application of modern machine learning methods capable of generating more accurate energy consumption forecasts for buildings to improve their energy efficiency. In this paper, based on the gradient boosting model, a method for modeling and forecasting energy consumption of buildings is proposed and computer algorithms for its software implementation in the SymPy computer algebra system are developed. To assess the efficiency of the proposed algorithms, a dataset that characterizes energy consumption of 300 commercial buildings is used. Results of computer simulations show that these algorithms improve the accuracy of energy consumption forecasts in more than 80% of cases as compared to other machine learning algorithms.
Similar content being viewed by others
REFERENCES
Gellings, C.W., The Smart Grid: Enabling Energy Efficiency and Demand Response, Fairmont, 2009.
Arghira, N., Hawarah, L., Ploix, S., and Jacomino, M., Prediction of appliances energy use in smart homes, Energy, 2012, vol. 48, no. 1, pp. 128–134.
Hawarah, L. and Jacomino, M., Smart home: From user’s behavior to prediction of energy consumption, Proc. 7th Int. Conf. Informatics in Control, Automation, and Robotics (ICINCO), Funchal, Portugal, 2010, vol. 1.
Breiman, L., Random forests, Mach. Learn., 2001, vol. 45, no. 1, pp. 5–32.
Breiman, L., Bagging predictors, Mach. Learn., 1996, vol. 24, no. 2, pp. 123–140.
Bergmeir, C. and Benitez, J.M., On the use of cross-validation for time series predictor evaluation, Inf. Sci., 2012, vol. 191, pp. 192–213.
Ediger, V. and Akar, S., ARIMA forecasting of primary energy demand by fuel in Turkey, Energy Policy, 2007, vol. 35, pp. 1701–1708.
Cincotti, S., Gallo, G., and Ponta, L., Modeling and forecasting of electricity spot-prices: Computational intelligence vs. classical econometrics, AI Commun., 2014, vol. 27, pp. 301–314.
Ardakani, F.J. and Ardehali, M.M., Novel effects of demand side management data on accuracy of electrical energy consumption modeling and long-term forecasting, Energy Convers. Manag., 2014, vol. 78, pp. 745–752.
Shchetinin, E.Yu., Cluster-based energy consumption forecasting in smart grids, Commun. Comput. Inf. Sci., Berlin: Springer, 2018, vol. 919, pp. 446–456.
Ireland's open data portal. https://data.gov.ie/dataset/energy-consumption-gas-and-electricity-civic-offices-2009-2012/resource.
Liaw, A. and Wiener, M., Classification and regression by random forest, R News, 2002, vol. 2, no. 3, pp. 18–22.
Friedman, J., Greedy function approximation: A gradient boosting machine, Ann. Stat., 2001, vol. 29, no. 5, pp. 1189–1232.
Burnham, K.P. and Anderson, D.R., Model Selection and Multi-Model Inference: A Practical, Information-theoretic Approach, Springer, 2002.
Kane, M.J., Price, N., and Scotch, M., Comparison of ARIMA and random forest time series models for prediction of avian influenza H5N1 outbreaks, BMC Bioinf., 2014, vol. 15, pp. 276–84. https://doi.org/10.1186/1471-2105-15-276
Granderson, J., Touzani, S., and Custodio, C., Accuracy of automated measurement and verification techniques for energy savings in commercial buildings, Appl. Energy, 2016, vol. 173, pp. 296–308.
Jain, R.K., Smith, K.M., Culligan, P.J., and Taylor, J.E., Forecasting energy consumption of multi-family residential buildings using support vector regression: Investigating the impact of temporal and spatial monitoring granularity on performance accuracy, Appl. Energy, 2014, vol. 123, pp. 168–178.
Zhou, Z.H., Ensemble learning, Encycl. Biom., 2015, pp. 411–416.
Zorita, A.L., Fernandez-Temprano, M.A., Garcia-Escudero, L.-A., and Duque-Perez, O., A statistical modeling approach to detect anomalies in energetic efficiency of buildings, Energy Build., 2016, vol. 110, pp. 377–386.
Amozegar, M. and Khorasani, K., An ensemble of dynamic neural network identifiers for fault detection and isolation of gas turbine engines, Neural Network, 2016, vol. 76, pp. 106–121.
Shchetinin, E.Yu., Melezhik, V.S., and Sevastyanov, L.A., Improving the energy efficiency of the smart buildings with the boosting algorithms, Proc. 12th Int. Workshop Applied Problems in Theory of Probabilities and Mathematical Statistics in the framework of Conf. Information and Telecommunication Technologies and Mathematical Modeling of High-Tech Systems (APTP+MS), 2018, vol. 2332, pp. 69–78.
Lyubin, P. and Shchetinin, E.Yu., Fast two-dimensional smoothing with discrete cosine transform, Commun. Comput. Inf. Sci., Berlin: Springer, 2016, vol. 678, pp. 646–656.
Gevorkyan, M.N., Demidova, A.V., Velieva, T.R., Korol’kova, A.V., Kulyabov, D.S., and Sevast’yanov, L.A., Implementing a method for stochastization of one-step processes in a computer algebra system, Program. Comput. Software, 2018, vol. 44, pp. 86–93.
GitHub Inc., SymPy. https://github.com/sympy.
SymPy Development Team. http://www.sympy.org/ ru/index.html.
Python official website. http://www.python.org.
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated by Yu. Kornienko
Rights and permissions
About this article
Cite this article
Shchetinin, E.Y. Development of Energy Saving Technologies for Smart Buildings by Using Computer Algebra. Program Comput Soft 46, 324–329 (2020). https://doi.org/10.1134/S0361768820050084
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0361768820050084