Прогнозирование динамики энергопотребления города Севастополя при использовании нейросетевых алгоритмов

D. Y. Kotelnikov; P. N. Kuznetsov; V. V. Kuvshinov; B. A. Yakimovich; A. M. Oleynikov

doi:10.22213/2413-1172-2021-1-78-86

Authors

D. Y. Kotelnikov Sevastopol State University, Sevastopol
P. N. Kuznetsov Sevastopol State University, Sevastopol
V. V. Kuvshinov Sevastopol State University, Sevastopol
B. A. Yakimovich Sevastopol State University, Sevastopol
A. M. Oleynikov Institute of Natural and Technical Systems, Sevastopol

DOI:

https://doi.org/10.22213/2413-1172-2021-1-78-86

Keywords:

forecasting, power consumption, distributed networks, methodology, neural network classification

Abstract

The global consumption of electric energy (EE) increases significantly every year, which leads to an increase in the interest of large consumers and electricity producers in such a procedure as forecasting energy consumption. The information obtained by forecasting can be used to appropriately distribute the current load in time and correctly calculate the perspective, so it becomes possible to reduce the duration of peak loads, which will improve the reliability of power consumption. Companies that produce EE will be able to build their economic development model more accurately and decide whether to expand the existing infrastructure due to the results of the forecast. It follows from the above that the forecast of energy consumption will reduce the financial losses of companies producing and consuming EE.

The paper proposes a method for identifying typical profiles of the daily dynamics of energy consumption, based on the methods of neural network analysis and neural network classification. The identified typical profiles later became the basis for the procedure of predicting energy consumption. Visually, the profiles were presented in the form of graphs of the daily dynamics of energy consumption, which allows you to visually and accurately assess the features of energy consumption for each of the feeders of the distribution network. The intermediate steps that were carried out to identify the typical profiles were also described. At the end, the energy consumption forecasting procedure was carried out, based on the use of the identified typical profiles, and its reliability was evaluated.

References

Чебоксаров В. В., Кузнецов П. Н. Пути решения проблемы нестабильности энерговыработки установок возобновляемой энергетики // Дневник науки. 2019. № 4 (28). С. 60–66.

Прогнозирование энергопотребления при использовании данных мониторинга электрических параметров / П. Н. Кузнецов, Д. Ю. Котельников, Д. Ю. Воронин, А. Г. Хомюк // Энергетические уста-новки и технологии. 2020. Т. 6, № 2. С. 71–76.

Ma Z., Yan R., Nord N. A variation focused cluster analysis strategy to identify typical daily heating load profiles of higher education buildings. Energy, 2017, no. 134, pp. 90-102. DOI: 10.1016/j.energy.2017.05.191.

Delzendeh E., Wu S., Lee A. The impact of occupants’ behaviors on building energy analysis: A research review. Renewable and Sustainable Energy Reviews, 2017, no. 80, pp. 1061-1071. DOI: 10.1016/j.rser.

05.264.

Hong T., Yan D., D’Oca S. Ten questions concerning occupant behavior in buildings: The big picture. Building and Environment, 2017, no. 114, pp. 518-530. DOI: 10.1016/J.BUILDENV.2016.12.006.

Haben S., Singleton C., Grindrod P. Analysis and clustering of residential customers energy behavioral demand using smart meter data. Energy & Buildings, 2019, no. 203, pp. 231-240. DOI: 10.1016/j.enbuild.

109455.

Rajabi A., Eskandari M., Ghadi M.J. A pattern recognition methodology for analyzing residential customers load data and targeting demand response applications. Energy & Buildings, 2019, no. 203, pp. 231-240. DOI: 10.1016/j.enbuild.2019.109455.

Yang T., Ren M., Zhou K. Identifying household electricity consumption patterns: A case study of Kunshan, China. Renewable and Sustainable Energy Review, 2018, no. 91, pp. 861-868.

Wang Y., Chen Q., Kang C. Clustering of Electricity Consumption Behavior Dynamics toward Big Data Applications. IEEE Transactions on Smart Grid, 2016, no. 7, pp. 2437-2447.

Zhou K., Yang C., Shen J. Discovering residential electricity consumption patterns through smart-meter data mining: A case study from China. Utilities Policy, 2017, no. 44, pp. 73-84. DOI: 10.1016/j.jup.2017.01.004.

Wang Y., Chen Q., Hong T. Review of smart meter data analytics: applications, methodologies, and challenges. IEEE Transactions on Smart Grid, 2019, no. 10, pp. 3125-3148. DOI: 10.1109/TSG.2018.2818167.

Drechny M. The method of consumers identification based on compressed power load profiles. IEEE in 2018 Innovative Materials and Technologies in Electrical Engineering, 2018, no. 18, pp. 55-58. DOI: 10.1109/IMITEL.2018.8370464.

Piscitelli M.S., Piscitelli M.S., Brandi S., Capoz-zoli A. Recognition and classiﬁcation of typical load proﬁles in buildings with non-intrusive learning approach. Applied Energy, 2019, no. 255, pp. 227-244.

Методика нейросетевого прогнозирования энергопотребления для устойчивого развития умной городской среды / П. Н. Кузнецов, Д. Ю. Котельников, Д. Ю. Воронин, А. Г. Хомюк // Дневник науки. 2020. № 5 (41). С. 30–44.

Кузнецов П. Н., Хомюк А. Г., Котельников Д. Ю. Методика оценки остаточного ресурса силового трансформатора // Дневник науки. 2020. № 5 (41). С. 44–59.