Hybrid Artificial Neural Networks Based Models for Electricity Spot Price Forecasting - A Review

Abstract

Electricity price forecasting plays a crucial role in a liberalized electricity market. In terms of forecasting approaches, artificial neural networks are the most popular among researchers due to their flexibility and efficiency in handling complexity and non-linearity. On the other hand, a single neural network presents certain limitations. Therefore, in recent years, hybrid models that combine multiple algorithms to balance out the advantages of a single model have become a trend. However, a review of recent applications of hybrid neural networks based models with respect to electricity price forecasting is not found in the literature and hence, the motivation of this paper is to fill this research gap. In this study, methodologies of existing forecasting approaches are briefly summarized, followed by reviews of neural network based hybrid models concerning electricity forecasting from year 2015 onwards. Major contributions of each study, datasets adopted in experiments as well as the corresponding experiment results are analyzed. Apart from the review of existing studies, the novelty and advantages of each type of hybrid model are discussed in detail. Scope of the review is the application of hybrid neural network models. It is found that the forecast horizon of the reviewed literature is either hour ahead or day ahead. Medium and long term forecasting are not comprehensively studied. In addition, though hybrid models require relatively large computational time, time measurements are not reported in any of the reviewed literature.