一类非凸约束二次规划问题的分支定界算法  

Branch-and-bound algorithm for non-convex constrained quadratic programming problems

作　　者：彭明丽 刘庆怀 张鸿洋 PENG Mingli;LIU Qinghuai;ZHANG Hongyang(School of Mathematics&Statistics,Changchun University of Technology,Changchun 130012,China)

机构地区：[1]长春工业大学数学与统计学院,吉林长春130012

出　　处：《长春工业大学学报》2024年第3期250-258,共9页Journal of Changchun University of Technology

基　　金：吉林省自然科学基金面上项目(20101597)。

摘　　要：针对一类非凸约束二次规划问题,提出一种新的参数化线性松弛分支定界算法,主要利用线性松弛技术求得原问题的全局最优值下界,以及区域删除规则缩减不可行区域,证明了算法的收敛性,最后通过数值实验表明算法的收敛速度加快,且该算法有效可行。The article introduces a novel parameterized linear relaxation branch-and-bound algorithm for a class of non-convex constrained quadratic programming problems.This algorithm primarily utilizes linear relaxation techniques tOobtain the global optimal lower bounds of the original problem and employs region deletion rules tOreduce the infeasible region.The convergence of the algorithm is proven,and numerical experiments demonstrate its accelerated convergence rate,indicating its effectiveness and feasibility.

关 键 词：分支定界 线性松弛技术 区域缩减 二次规划 

分 类 号：O224[理学—运筹学与控制论]