Hybrid filter-wrapper feature selection for short-term load forecasting

13 Selection of input features plays an important role in developing models for short14 term load forecasting (STLF). Previous studies along this line of research have focused 15 pre-dominantly on filter and wrapper methods. Given the potential value of a hybrid 16 selection scheme that includes both filter and wrapper methods in constructing an 17 appropriate pool of features, coupled with the general lack of success in employing filter 18 or wrapper methods individually, in this study we propose a hybrid filter-wrapper 19 approach for STLF feature selection. This proposed approach, which is believed to have 20 taken full advantage of the strengths of both filter and wrapper methods, first uses the 21 Partial Mutual Information based filter method to filter out most of the irrelevant and 22 redundant features, and subsequently applies a wrapper method, implemented via a firefly 23 algorithm, to further reduce the redundant features without degrading the forecasting 24 accuracy. The well-established support vector regression is selected as the modeler to 25 implement the proposed hybrid feature selection scheme. Real-world electricity load 26 * Corresponding author: Tel: +86-27-87558579; fax: +86-27-87556437. Email: [email protected] or [email protected]