一个参数动态调节的全局凸填充函数算法

A Globally Convexized Filled Function Algorithm with an Adjustable Parameter

构造了有界闭箱上连续全局优化问题的一个新的全局凸填充函数,分析了该函数的几个性质,设计了一个基于该填充函数的全局优化算法。该算法通过动态调节参数来跳出当前收敛的局部极小解的邻域,数值试验表明该算法是有效的。

In this paper, we present a new globally convexized filled function for the box constrained continuous global minimization problem. We propose an algorithm for finding a global minimizer of the box constrained continuous global minimization problem, by minimizing the filled function. The algorithm can escape successfully from a previously converged local minimizer by taking the value of parameter increasingly. Numerical experiments on six standard testing problems show that the algorithm is effective.