Support Vector Regression with Multi-Strategy Artificial Bee Colony Algorithm for Annual Electric Load Forecasting

摘要

A novel support vector regression (SVR) model with multi-strategy artificial bee colony algorithm (MSABC) is proposed for annual electric load forecasting. In the proposed model, MSABC is employed to optimize the punishment factor, kernel parameter and the tube size of SVR. However, in the MSABC algorithm, Tent chaotic opposition-based learning initialization strategy is employed to diversify the initial individuals, and enhanced local neighborhood search strategy is applied to help the artificial bee colony (ABC) algorithm to escape from a local optimum effectively. By comparison with other forecasting algorithms, the experimental results show that the proposed model performs higher predictive accuracy, faster convergence speed and better generalization.