Particle swarm optimization(PSO) is a kind of stchastic and global search algo-rithm based on the behavior and intelligence simulation of biology swarm. Because of its inherent parallelization, robustness, simple operation and less parameters, PSO was received extensive attention, and widely applied in the fields of science, engineering and so on. When applied to high dimensional and complex problems, this algorithm was easy to fall into local optimum and pre-convergence. Thus this paper proposes Double Evalutionâ€™s Particle Swarm Optimization and discusses its application.Contrary to Standard Particle Swarm Optimization doesnâ€™t consider change of current particleâ€™s location and fitness in iterative process, so some particlesâ€™indi-vidual cognitive abilities may not improve sometime. Firstly, the change of current particleâ€™s location and fitness is compared with iterated particleâ€™s in Double Evalu-tionâ€™s Particle Swarm Optimization. Then. utilizing Genetic Algorithmsâ€™mutation thinking particle which is worsed than iterated particleâ€™s location and fitness imu-tated, in order to improve particlesâ€™ individual cognitive abilities. Finally, utilizing eight, test functions tests algorithm. experimental results show that Double Evalu-tionâ€™s Particle Swarm Optimization can accelerate speed of convergence, but improve solution accuracy.The knapsack problem is typical combinatorial optimization problem, it is ap-plied in capital budgeting, resource allocation practical problems and so on. This paper solves the0-1knapsack problem by Double Evalutionâ€™s Particle Swarm Op-timization. Solving the0-1knapsack problemâ€™s instance results show that Double Evalutionâ€™s Particle Swarm Optimization can effectively solve combinatorial opti-mization problem. |