Abstract:
The nationally declared "Energy Strategy of Russia for the Period up to 2035" involves a
comprehensive structural transformation of the energy sector and its transition to a qualitatively new level
that meets the needs of the country's economic development. Improvement of energy efficiency is one of
the priority directions of the policy outlined in Russian energy strategy for this period. Nowadays steps are
being taken to encourage consumers to participate in increasing energy efficiency and leveling load
schedules. The need for highly accurate forecasts of both energy consumption and production is becoming
increasingly important. Moreover, a growing number of enterprises are becoming interested in improving
their own energy efficiency and reducing electricity consumption costs, thereby increasing their profits.
This research aims to develop a highly accurate mathematical model for forecasting electricity production
volume. The accuracy of the model is enhanced by incorporating only dominant factors into the model,
with a preliminary determination of the type of their functional dependence on the studied parameter.
Additionally, these factors are shifted back by several periods to use reliable values of all factors in the
model and avoid constructing separate forecasting models for them. This approach prevents the forecast
error of each factor from compounding the overall model error. The magnitude of the shift is justified by
the autocorrelation function.
Keywords:improvement of energy efficiency, forecasting model, multifactor regression, determination
of functional relationship, identification of dominant factors, demand management, electric power industry,
planned production volume, day-ahead market, minimum equilibrium price index