基于改进EGO算法的黑箱函数全局最优化

Global optimization of black-box function using improved EGO algorithm

基于Kriging模型的EGO算法是一种适用于黑箱函数求极值的全局最优化算法,但该算法忽略了对Kriging模型精度的控制。针对该算法的不足之处,提出了兼顾Kriging模型精度与模型寻优的迭代函数,并将改进后的EGO算法应用于五个检验函数及一个存货模型,从Kriging模型精度及优化结果两方面对改进前后的算法进行比较。结果表明,改进后的EGO算法提高了最终Kriging模型的精度,并在对目标函数进行少量估值的情况下获得了更为全局化的最优解。

EGO algorithm based on Kriging model is a suitable method for the global optimization of black-box function, but it ignored the accuracy of Kriging model. To overcome the shortcoming of EGO algorithm, this paper proposed an improved algo- rithm, and it＇ s iterative function took into account the accuracy and the optimization of the Kriging model. Then this paper ap- plied the algorithm to five test functions and an inventory model. The results show that compared to the original EGO algo- rithm, the improved algorithm can improve the final accuracy of the Kriging model and obtain a more globally optimal solution via a small amount of the valuations to the objective function.