Research On The Filled Function Methods For Global Optimization Problems

The filled function method is one of the methods for solving global optimization problem.This method only combine the classic method with the filled function,so it has been popular by many researchers.This paper studied the filled function method of unconstrained global optimization problems.This paper is divided into four chapters.In the first chapter,we briefly introduce the basic concepts of global optimization and the filled function method.Based on this,propose the main purpose and work of this paper.In the second chapter,two filled functions with one parameter are designed.Then corresponding filled function algorithms are established and numerical results are given to show that filled function method are effective for solving global optimization problems and algorithm II is better than algorithm I.In the third chapter,two new filled functions without parameter are established.Then algorithms are designed with its theoretical nature.Finally,through the numerical experiments,algorithms are proven to be efficient and algorithm II is better than algorithm I.The last chapter make conclusions of this paper and the development direction of filled function method is prospected.