提出一种新的求解无约束全局优化问题的方法,该方法把修正的BFGS方法与填充函数方法相结合.使得目标函数f(x)的当前局部极小点x1可以移到目标函数的另一个局部极小点x.且f(x1)≥f(x),同时i也是填充函数的极小值点;然后再以x为初始点求f(x)的局部最优解.反复以上过程,最终可以找到f(x)的全局最优解.
A new method for finding a global solution of the unconstrained optimization was introduced. The algorithm combines the modified BFGS with the filled function so that it can make the present local minimize x1 transfer to another local minimize x under the condition f(x1 )≥f(x). Starting at the new initial point x and repeating the above process several times,we can find the global solution of the unconstrained optimization.