FORECASTING METHOD AND SYSTEM OF DEMAND FOR ELECTRIC POWER

摘要

The present invention relates to a method and a system for predicting electric power demand. The method for predicting an electric power demand in accordance with the present invention comprises: (a) a step where a temperature probability distribution function estimation unit estimates monthly and weekly national temperature probability distribution functions by estimating monthly and weekly temperature probability distribution functions using local temperature data and placing a weighted value depending on sales volume for each area; (b) a step where a reaction function estimation unit estimates a temperature reaction function for monthly sales volume and weekly maximum electric power in which a change in load patterns in accordance with a period is considered and a relative energy price reaction function which compares other energy prices with an electricity rate index; (c) a step where a temperate effect estimation unit estimates a temperature effect of the monthly sales volume and the weekly maximum electric power using the temperature probability distribution function, the temperature reaction function and the relative energy price reaction function; (d) a step where a prediction model estimation unit estimates a prediction model of the monthly sales volume and the weekly maximum electric power comprising a polynomial expression which includes a time change coefficient so as to be changed as time passes using the estimated temperature effect; and (e) a step where a prediction model testing unit predicts a model by testing the prediction model using a co-integration model or an error correction model.