Combined models of electricity consumption

Abstract

The article investigates models and methods of electric load forecasting. It is shown that among the known methods of power consumption management, preference is given to those based on the use of forecast estimates. The analysis of works devoted to the issues of forecasting the processes of power consumption management systems of industrial enterprises is carried out. It is shown that it is expedient to use adaptive models as a basis for operative forecasting of loads of power supply systems of industrial enterprises. Analysis of adaptive models of electricity consumption forecasting based on the method of exponential smoothing showed their high efficiency and good adaptability to changes in the process of electricity consumption. It is shown that the greatest difficulty in forecasting are cases of abrupt changes in the development of the process. Abrupt changes in the process can lead to a violation of pre-existing qualitative relationships of the parameters of the projected system. If the jump is the transition of the predicted system from one steady state to another, the model of exponential smoothing with correction of the constant smoothing has the best adaptability to this kind of change. At the same time, changes of the "pulse" type are worked out by the model with a certain delay, which leads to an increase in the standard error of the forecast. Therefore, the model's response to change slows down. To eliminate this circumstance, a forecasting procedure based on combined models is proposed. The paper considers two models of combined forecasting - a combined model of joint processing of forecasting results and a combined model of selective type. Experimental studies of the considered models are carried out.