中文摘要：

准确的风电出力预测对电力系统的安全稳定运行和减少系统运行成本至关重要。将BN分解法、蚁蛳优化算法（ALO）和最小二乘支持向量机模型（LSSVM）相结合,提出了一种短期风电出力预测BN-ALO-LSSVM混合模型。该模型首先将风电出力原始时间序列分解为各子序列,进而运用LSSVM模型对各子序列分别进行预测;与此同时,为提升预测精度,运用ALO群体智能优化算法确定LSSVM模型的最优参数。实例结果表明：与LSSVM,BN-LSSVM和ALO-LSSVM模型相比,本文提出的风电出力预测BN-ALO-LSSVM混合模型的预测精度最高,且是有效可行的。

英文摘要：

Accurate wind power forecasting is vital to the safe and stable operation of electric power system and the system operation cost reduction. The BN decomposition method, Ant lion optimizer algorithm and Least Square Support Vector Machines are combined, and a new hybrid wind power forecasting model is proposed, namely BN-ALO-LSSVM model. In this proposed model, the original wind power time series are firstly decomposed into different subsequences, and then they will be respectively forecasted by using LSSVM model. In order to improve the forecasting accuracy, the parameters of LSSVM model are optimally determined by employing a new swarm intelligent algorithm ALO. The empirical results show that compared with LSSVM, BN-LSSVM and ALO-LSSVM models, the proposed BN-ALO-LSSVM model has the best forecasting performance, and the forecasting accuracy is the highest. The proposed BN-ALO-LSSVM model for wind power forecasting is effective and feasible.