Comprehensive approach to modeling electricity consumption and identifying patterns using data mining methods

¹ Reshetnev Siberian State University of Science and Technology, 660037 Krasnoyarsk, Russia
² Bauman Moscow State Technical University, 105005 Moscow, Russia
³ Moscow Timiryazev Agricultural Academy, Russian State Agrarian University, 127550 Moscow, Russia

^* Corresponding author: vadimond@mail.ru

Abstract

In this study, a detailed analysis of a dataset related to electricity consumption under various conditions was conducted. The dataset includes comprehensive information such as indoor temperature at the time of data recording, humidity levels, room size, number of people present, use of heating, ventilation, and air conditioning (HVAC) systems, lighting, day of the week, and the presence of holidays. This extensive coverage of factors provides an opportunity for a deep understanding of the relationships between electricity consumption and different variables. Several significant patterns were identified during the analysis. One key finding was the detection of a connection between temperature readings and electricity consumption, confirming the importance of temperature conditions in energy usage. A correlation was also found between energy consumption, HVAC system use, and humidity levels, indicating the substantial impact of these factors on overall electricity consumption. Other factors such as lighting use, holidays, renewable energy sources, and the day of the week also proved significant during factor analysis. These results were further validated using Self-Organizing Maps (SOMs), which demonstrates the reliability of the identified patterns. However, when attempting to model the data, the results were less satisfactory. The models showed only marginal improvement when combining methods, suggesting the need for further work. To achieve more accurate and reliable results, it is recommended to supplement the dataset with additional information or apply more sophisticated analytical methods. These steps will help improve model quality and more effectively utilize the identified patterns for forecasting and optimizing electricity consumption.