Filled Function Methods For Constrained Global Optimization Problems

This paper presents a new filled function method to solve global op-timization problems with both box and inequality constraints.In Chapter1 we introduce the applied background and main techniques of global op-timization problem,particularly,the filled function method is emphaticallymade an introduction.In Chapter 2,we study the continuous constrainedglobal optimization problems,a new filled function is given,some proper-ties of the filled function and corresponding algorithm is discussed. Thethird chapter focuses on the filled function method of integer programmingwith box and nonlinear inequality constraints. We propose a novel filledfunction for the constrained integer programming problem with inequalityconstraints based on filled function methods proposed in [12] and [35].More-over,some properties of the filled function are discussed and proved.A filledfunction algorithm is designed to obtain a discrete global minimizer of non-linear integer programming problem with both inequality constraints re-ferred to algorithms proposed in [12], the computational results show thatthis algorithm is e?cient and reliable.