基于混合核函数FOA-LSSVM的预测模型

支持向量机（SVM）的核函数类型和超参数对预测的精度有重要影响。由于局部核函数学习能力强、泛化性能弱，而全局核函数泛化性能强、学习能力弱的矛盾，通过综合两类核函数各自优点构造了基于全局多项式核和高斯核的混合核函数，并引入果蝇优化算法（FOA）对最小二乘支持向量机（LSSVM）参数进行全局寻优，提出了混合核函数FOA-LSSVM 预测模型。结果表明，该模型较传统方法在电力负荷预测精度上有了明显提高，预测结果科学可靠，在预测中具有良好的实际应用价值。

Forecasting model via LSSVM with mixed kernel and FOA

The kernel and the parameters of Support Vector Machine（SVM） have a significant impact on precision. In view of better learning capability of local kernels and better generalization capability of global kernels, the mixed kernel is constructed by a typical local kernel-Radial Basis Function（RBF） and a typical global kernel-polynomial kernel. By use of Fruit Fly Optimization Algorithm（FOA）, a novel FOA-LSSVM model with mixed kernels is set up in this paper. Results demonstrate that the new model has great accuracy than traditional methods and has real application value in forecasting.