Rethinking weather station selection for electric load forecasting using genetic algorithms

Abstract

Demand forecasting is and has been for years a topic of great interest in the electricity sector, being the temperature one of its major drivers. Indeed, one of the challenges when modelling the load is to choose the right temperature, or set of temperatures, for a given load time series. However, minimum research efforts have been devoted to this topic. This paper reviews the most relevant methods that were applied during the GefCom2014 electricity demand forecasting competition and presents a new approach to weather station selection, based on Genetic Algorithms, which allows finding the best combination of temperatures for any demand forecasting model, and outperforms the results of those methods. In addition, our empirical results show that using different weights to combine the temperatures, optimized using the Broyden Fletcher Goldfarb Shanno algorithm, allows achieving significant error improvements.

selection and combination, Genetic Algorithms, BFGS algorithm

Demand forecasting is and has been for years a topic of great interest in the electricity sector, being the temperature one of its major drivers. Indeed, one of the challenges when modelling the load is to choose the right weather station, or set of stations, for a given load time series. However, only a few research papers have been devoted to this topic. This paper reviews the most relevant methods that were applied during the Global Energy Forecasting Competition of 2014 (GEFCom2014) and presents a new approach to weather station selection, based on Genetic Algorithms (GA), which allows finding the best set of stations for any demand forecasting model, and outperforms the results of existing methods. Furthermore its performance has also been tested using GEFCom2012 data, providing significant error improvements. Finally, the possibility of combining the weather stations selected by the proposed GA using the BFGS algorithm is briefly tested, providing promising results.