为求解广义几何规划问题,提出一个新的线性化松弛技巧.在此基础上,给出一个新的分支定界算法.为进一步改进算法,构造一个新的删除技巧,该技巧可被用来提高算法收敛效率.理论上证明了算法的收敛性,数值试验显示本文方法是有效可行的.
In this paper,based on a new linearization method,a new branch and bound algorithm for globally solving generalized geometric programming is presented. Furthermore, a new pruning technique is proposed, which can be used to improve the convergence speed of this algorithm. Con- vergence of the algorithm is shown, and some numerical experiments are given to illustrate the feasi- bility and efficiency of the algorithm.