Grid-based simulation method for spatial electric load forecasting using power-law distribution with fractal exponent

Abstract

A grid-based simulation method to forecast the spatial growth of load density in a distribution utility service zone is presented. The future load density is simulated considering a city's dynamic growth. A power-law distribution with fractal exponent is used to determine how much the load density will increase/decrease depending on different factors like: natural growth following patterns, spontaneous new loads without predefined patterns, and urban poles growth that attract/repel new consumers. These factors are simulated using two modules that exchange information. The final result will display a map with future load density considering a heterogeneous distribution, which is the main contribution of this work. The method is tested on a real distribution system in a medium-sized Brazilian city. The most important characteristic of this method is its high efficiency in spatial load density recognition with less input data than traditional spatial load forecasting methodologies. Copyright © 2015 John Wiley & Sons, Ltd.