This paper describes a model to forecast the daily maximum demand of Malaysian large steel mills and the annual maximum demand contributed by these steel mills. This study attempts to combine both the top-down and bottom-up approaches to forecast the daily and annual maximum demand of the steel mills. The top-down approach uses regression analysis to forecast the annual amount of electricity consumption of the steel mills. The bottom-up approach uses the Model for Analysis of Electric Demand_Electric Load (MAED_EL) to convert the annual steel mills electricity consumption (which was earlier obtained from the regression model) into hourly load of the steel mills. The proposed method shows good forecasting accuracy, with weekly Mean Absolute Percentage Error (MAPE) of 2.3%.
Keywords: Load forecasting, regression analysis, top-down, bottom-up, steel mills.
Copyright © 2018 Author(s) retain the copyright of this article.
This article is published under the terms of the Creative Commons Attribution License 4.0