Tabu Search Based Bi-Group Particle Swarm Optimization And Application Research

Particle Swarm Optimization(PSO) is inspired by social behavior of bird flocking or fish schooling and was originally introduced by Kennedy and Eberhart in 1995.It is a new kind of evolutionary computation and a population-based,self-adaptive search optimization technique.Pso is simple in concept,few in parameters,and easy in implementation.As a kind of swarm intelligence,it has proven to be a powerful global optimization method.PSO has been widely applied in function optimization and shows great potential in practice.Now,it has been widely applied in many other areas,such as function optimization,artificial neural network and fuzzy system control.In this paper,the basic principles of PSO are introduced.The research progress on PSO algorithm is summarized such as convergence analysis,parameter selection,two standard PSO algorithms(inertia weight model and constriction factor model),application of PSO algorithm etc.The Particle Swarm Optimization has some demerits of slow convergence velocity and easily being relapsed into local extreme, and the inertia weights also impose influence on the searching capability of particle swarm at different time. Condidering the different demand on exploring and searching ability of the particle swarm at each phase, this paper proposes a noval Tabu Search based bi-group Particle Swarm Optimization(TSBBPSO).It combines with the influence of initial solution domain in searching the optimum solution. This algorithm takes full use of the short-time and long-time memory ability of the Tabu Searching.It devides the particle swarm into two different sub-groups which perform simultaneously. There are more particles of high inertia weights in the prophase one in order to make the large-scale searching in solution domain easy. While the inertia weights of particles in the anaphase one descend in linearity which improve the ability of partial searching. With the variety of the particles quantity of the two sub-groups and the influence of the inertia weights, it can realize a better searching ability in both partial and overall situations. The exchange and amalgamation of the information can also be realized through the subgroups'recombination. Adopting the Tabu Searching after iterative algorithms can not only effectively solves the demerit that the tabu search method relies on the initial solution too much, but also takes good use of its hill-climbing capability to avoid the particle swarm algorithm getting into partial optimization too rapidly.Meanwhile, the searching in the neighborhood of the possible optimal solution domain with the long-time memory capability of the tabu search method equals to a purposeful variation, and it improved the algorithm's capability of searching the optimal solution in the overall domain. The effectiveness of this algorithm is demonstrated for constringency rate and accuracy through a number of typical computational experiments.In order to solve packing problem with improved PSO, we first analyse the computational complexity of packing problem. After that we describe the rectangle packing problem and modeling method in detail. Then we make use of the TSBBPSO to solve the problem. And then we also reach a good result by using the TSBBPSO to solve the unequal circle problem which is of high complexity. It also reaches a good result. At last,we use the TSBBPSO in a special example of polygon packing problem-unit equilateral triangles packing problem. Therefore,the TSBBPSO is a good way to solve the packing problem.