A Holistic Feature Selection Method for Enhanced Short-term Load Forecasting of Power System
Bozhen Jiang, Yi Liu, Hua Geng, Yidi Wang, Huarong Zeng, Jiangqiao Ding
Abstract
Short-term load forecasting (STLF) is important for the operational security and economics of power system. However, most of STLF methods lack an efficient feature selection method to model the time series nonlinearities and feature interaction. In this paper, a new holistic feature selection method is presented. The feedforward long short-term memory network (F-LSTM) is proposed to learn the nonlinear mapping function between features and load. Then, a feature importance matrix is designed to reflect relevancy, redundancy and interaction among the candidate features. Moreover, a hybrid filter-wrapper approach is developed to select suitable features efficiently. The filter part separates useless information for the trained F-LSTM output. The wrapper part selects the optimal subset by fine-tuning the threshold. The results from an empirical study in Switzerland suggest that: 1) the selected subset of features shows high relevancy, low redundancy and high interaction, which is also consistent with the features selected by other feature selection methods, and 2) the proposed method has good prediction performance and can be applied to various artificial neutral network based short-term load forecasting models, which delivers an average 12.1% improvement.