A New Solution Algorithm for General Constrained Twice- Differentiable NLP Problems
In this paper, new deterministic global optimization methods are proposed, that are applicable to univariate and multivariate nonlinear programming problems respectively composed of twice differentiable objective and constraint functions. Both methods employ a difference of convex underestimator, that is a concave quadratic function which does not need an iterative local optimizer to determine the lower bounding value of the objective function and a convex cut function which effectively cuts infeasible regions for nonconvex constraints. Both methods are proven to have a finite ε- convergence to locate the global optimum point.
Global optimization Difference of convex underestimator Convex cut function
Young Cheol Park Min Ho Chang Tai-yong Lee
Department of Chemical & Biomolecular Engineering,Korea Advanced Institute of Science and Technology,373-1,Guseong-dong,Yuseong-gu,Daejeon 305-701,Korea
国际会议
西安
英文
2007-08-15（万方平台首次上网日期，不代表论文的发表时间）