Learning algorithm of artificial neural network factor forecasting power consumption of users

Abstract

Seasonal fluctuations in electricity consumption, an uneven load of supply lines reduce not only the indicator of energy efficiency of networks but also contribute to a decrease in the service life of elements of power supply systems. Revealing the patterns of such fluctuations makes it possible to build models of power consumption, predict its dynamics, which in general will contribute to ensuring the energy efficiency of urban electrical networks and increasing the reliability of power supply systems. A computational, computer and neural network model is proposed that allows to increase the accuracy of the forecast of electricity consumption by household consumers. Based on the developed mathematical model, taking into account the obtained factor coefficients - ti, h, c, s, k for 2020 for 9 cities of the Republic of Tajikistan, monthly coefficients characterizing the terrain conditions (αi)  were calculated. The results obtained using the proposed method was compared with the results of a computer and neural network model.