Forecasting consumption of electric energy by using wavelet transform

Abstract

Forecasting electricity consumption is an urgent task for generating companies, since it is currently impossible to accumulate electricity on an industrial scale. Also, the forecast is necessary for consumers to carry out technical work and other activities. The purpose of this work is to make a forecast of electric energy consumption using the wavelet transform, and to select the optimal wavelet function for forecasting. Data for forecasting is a schedule of the load of the shop, which plays the role of a household room, warehouse, as well as a working office for personnel who service electrical installations at a production enterprise. Based on the results of the work, the optimal wavelet function is selected. The result of the work is a representation of the trend of electric energy consumption by the object under consideration, i.e. a forecast presented in the form of a graph, and a detailed component of the projected consumption is obtained, which in theory is justified as interference and sharply variable nature of electricity consumption.

Keywords

• short-term forecasting
• time series
• mathematical model of forecasting
• analysis of electricity consumption
• wavelet transform
• load graph
• forecasting methods