Predicting Load Suitability Requires Learning Customer Behavior

Abstract

Due of its significance for the smart grid, load forecasting has been the subject of extensive research. There are several customer kinds and varying energy consumption patterns in the existing Smart Grid. Customer behaviors refer to a customer's habits of energy usage. If consumer habits could be taken into account, load forecasting in a grid would be greatly aided. In an effort to improve the grid's overall load forecasting accuracy, the aforementioned study suggests a novel technique that groups various customer classes according to their observed behaviors and calculates the load of each customer cluster.