Particle Swarm Optimization Algorithm And The Application In Multi-objective Optimization

Multi-objective optimization problem is the common kind of optimization problems in our real life. Its main characteristic is there may be some conflicts between all the objectives, so it isn’t possible to make all the goals achieve the optimal solution, namely a solution may be good for a goal, but inferior for other goals. All of these solutions form a pareto optimal solution set of the multi-objective optimization problems. The solving methods of traditional multi-objective optimization problem have some advantages, such as simple, efficient, can converge to a pareto optimal solutions and so on, but it still has many defects. Therefore, looking for more efficient optimization algorithm has become one of the research goals of scholars.Particle swarm algorithm is a population intelligent evolution algorithm developed in recent years. Since the particle swarm algorithm was proposed, it getted extensive attention of the scholars because of its advantages of simple concept, easy implementation, etc. And it have been widely used in image processing, neural network, objective optimization and so on. In this paper, basic principle of the particle swarm optimization algorithm, parameter setting, algorithm improved and the application in multi-objective optimization are researched, and the main working and innovation points are as follows:1. The basic principle of the particle swarm algorithm, algorithm process, parameters settings and so on are analysised. And some common methods of improving particle swarm optimization algorithm at present are summarized. In addition, the particle swarm algorithm and genetic algorithm are analysised contrastly.2. Basic concepts of the multi-objective optimization problem introduction is introduced systematically, and the advantages and limitations in the traditional multi-objective optimization methods are analyzed in detail. In addition, several common resolving methods of traditional multi-objective optimization problems and solutions based on the evolutionary algorithm are summarized.3. Contraposing problems of the premature convergence in the standard particle swarm algorithm, a kind of improving methods of the particle swarm algorithm based on chaos thought is put forward. As for inertia particles sinked into the local extreme value point, methods of the chaotic sequence initialized again is adoptted to help the inert particles escape from bondage, and the global optimal solution is seeked out quickly.4. The particle swarm algorithm can search for more than one solution in the solution space hidden parallelly, and it can improve the efficiency of concurrent solving through the similarity between the solutions, so it is very suitable for solving the multi-objective optimization problem. In this paper, optimal solution evaluation selection and chaos thought are introduced to the multi-objective particle swarm algorithm, a kind of chaos multi-objective particle swarm algorithm based on the optimal solution evaluation selection is proposed.