ІНТЕРВАЛЬНЕ ПРОГНОЗУВАННЯ ПРОЦЕСУ ЕЛЕКТРОСПОЖИВАННЯ

The paper considers the solution of the problem of interval forecasting of electricity consumption. It is shown that the following criteria are used to evaluate the quality of short-term electricity consumption forecasting: forecasting accuracy; interval; resistance to data errors; model adaptability. It is shown that the quality of a forecasting model is largely determined by its accuracy, which is a measure of the adequacy of the future process with its assessment. It is shown that observation of the real process is carried out under conditions of interference, and the probability of obtaining an absolutely accurate forecast is quite low. In this regard, the range of predicted values of the electricity consumption process is introduced. A procedure for interval forecasting of the electricity consumption process for a linear model is proposed, which is based on determining the variance of the forecast error.