Research On Continued Explosion Algorithms And Invex Optimization Algorithms

Intelligent Optimization Algorithm is a very active optimization algorithm in recent years. As its extensive application, researchers in all fields have more and more attentions to the Intelligent Optimization Algorithm. Compared with the Classical Algorithms, it not only can realize easily in practical application, but also can achieve global convergence well in solving physical problems. The ideas of Intelligent Optimization Algorithm converts method orientation to problem orientation, so a series of Intelligent Optimization Algorithm are proposed one after another,such as Explosion Search Algorithm with Neighborhood Search Mechanism, Grenade Explosion Method. In this paper, according to the idea of bomb explosion, a new Intelligent Optimization Algorithms called Continued Explosion Algorithm is proposed.There are three parts in this paper:First, it introduces the extensive application of optimization problem, describes the background and research status of some algorithms and gives the basic steps and process diagram of some algorithms.Second, a new Intelligent Optimization Algorithms called Continued Explosion Algorithm is proposed. It gives the basic ideas, the mechanism, the realization ways, the basic steps and the process diagram of algorithm.Through the test of the standard function and a large number of experiment results, it also confirms the availability and practicability of the algorithm. Then, according to the short of algorithm,it proposes the improved measures and confirms the superiority of algorithm by the test of the standard function with the improved algorithm. Finally, combined the improved algorithm with the fininsearch and fminunc functions in the toolbox, it gets a better fitness value.Thirdly,in real vector spaces, the (F, K)-invex sets is the extension of the E-invex sets. In this paper, an optimization algorithm based on one-dimensional search is proposed to solve the invex optimization problems whose constraint set is the (F, K)-invex sets. The computing result of the new optimization algorithm is more remarkable than that from the functions of the Optimization Toolbox. The proposed optimization algorithm also provides a new idea for the improved nonlinear optimization algorithm.