摘要
本文提出了一种新的适用于处理非线性约束下线性规划问题的全局优化算法。该算法通过构造子问题来寻找优于当前局部最优解的可行解。该子问题可通过模拟退火算法来解决。通过求解一系列的子问题,当前最优解被不断地更新,最终求得全局最优解。最后,本算法应用于几个典型例题,并与罚函数法相比较,数值结果表明该算法是可行的,有效的。
A new global optimal algorithm is presented to solve linear programming problems with nonlinear constraints. In this algorithm, a subproblem is set up to search for a new feasible point at which the value of the objective function is lower than the current local minimum. The subproblem is solved by simulated annealing algorithm. Through solving a sequence of subproblems, the current optimal feasible solution can be incessantly renewed and the global optimal solution can be got at last. Finally, this algorithm is applied to some test problems and it has the better results than using the penalty function algorithm.
出处
《运筹与管理》
CSCD
2007年第1期28-31,共4页
Operations Research and Management Science
基金
辽宁省教育厅科学研究计划资助项目(2005040)
关键词
运筹学
非线性规划
模拟退火
非线性不等式约束
operations research
nonlinear programming
simulated annealing
nonlinear inequality constraints
作者简介
钱伟懿(1964-),男,辽宁省锦州市人,教授,博士,硕士生导师,主要从事:优化理论与应用、智能优化算法等方向研究;
杨宇(1981-),女,辽宁省大石桥市人,硕士研究生,从事优化理论与应用的研究。