Modernization of the method for predicting energy consumption of minigrid load nodes by using extrapolation of the trend line obtained using wavelet transformation of the electrical load graph

Subject of research: methods for predicting electrical load schedules. Purpose of research: improvement of the forecasting method for assessing the growth of minigrid energy consumption. Methods and objects of research: elements of the theory of wavelet transform of time series, approximation methods were adopted as a research method. Main results of research: as a result of the study, the method for predicting the load power of an isolated low-power electrical system (minigrid) was modernized; as an element of modernization, the use of wavelet transforms for filtering the high-frequency component of the power consumption graph is proposed. The paper also discusses the basic principles of operation of wavelet transforms, their advantages and possibilities of application in forecasting problems, and also provides an example of the successful use of this method.