摘要
针对凸约束不定二次规划问题,给出一个分枝界定方法。通过将凸约束不定二次规划问题等价地转化为凸凹规划问题,利用超矩形体的二分技术和锥剖分技术,在超矩形体上确定原问题的最优解,并进行了收敛性分析。
A brance-and-bound method is provided for the indefinite quadratic programming problems under convex constraints. By equally converting the indefinite quadratic programming problems under convex constraint into convex concave problems, and by using super rectangle dissection skill and cone dissection skill, the optimal solution to the original problems is determined from the super rectangle,and the convergence is analyzed.
出处
《渤海大学学报（自然科学版）》
CAS
2007年第2期166-168,共3页
Journal of Bohai University:Natural Science Edition
基金
辽宁省教育厅基金资助项目(No:2005040)
关键词
不定二次规划
凸凹规划
线性规划
分枝定界方法
锥剖分
整体优化
indefinite quadratic programming
convex concave programming
linear programming
branch-and-bound method
cone dissection
global optimization
