International Scientific Journals
of Scientific Technical Union of Mechanical Engineering "Industry 4.0"

This report presents an approach for predicting energy consumption in a specific living space based on experimental data on the indoor microclimate. The data for the analysis of the temperature in the room and the external temperature of the environment is used for the period from 01.01.2016. until 31.12.2023, and a period of 18:00 – 24:00. Regression analysis is used to create a model for increasing energy consumption.

This report analyzes the accuracy of predicting by the method of least squares and neuron networks. This study compares the errors and deviations of the predicted results from the actual ones. Considered are three options – when the parameter is defined by linear, quadratic and exponential law.