Research On Particle Swarm Optimization And Its Applications

Recently, simulating the behavior of biological colony to solve computation problems is becoming a hot research topic and a new academic framework based on Swarm Intelligence formed accordingly. Particle Swarm Optimization (PSO) algorithm is based on swarm intelligence theory. The algorithm can provide efficient solutions for optimization problems through intelligence generated from complex activities such as cooperation and competition among individuals in the biologic colony. This work concentrated on PSO based algorithm and its potential engineering applications such as neural network training, constraint optimization and combinatorial optimization, and then presented its potential of these applications.Firstly, particle swarm optimization is introduced and its developments are reviewed. The basic applications of PSO algorithm and its engineering applications are summarized. The future research directions of PSO algorithm are pointed out and its potential applications are presented.Secondly, Structure-improving Particle Swarm Optimization (SPSO) algorithm for training artificial neural network (ANN) is proposed. By tuning the structure and connection weights of ANN simultaneously, the proposed algorithm eliminates some ill effects introduced by redundant input features and the corresponding redundant structure of ANN, and obtains optimized ANN with problem-matched capacity for information processing. The ANN trained by SPSO is applied for water quality classification, prediction and credit evaluation. Comparing with traditional training algorithms such as BP and genetic algorithm, SPSO demonstrated its high efficiency and robustness.Thirdly, particle swarm optimization based algorithm is used to solve load dispatch problem and tolerance allocation problem. Constraint handling strategy suit for PSO mechanism is proposed. Furthermore, through combination with direct search, the search ability of PSO algorithm in local region is intensified. The efficiency of this algorithm is validated by the experiments. The proposed algorithm could be applied for any engineering optimization application that can be mathematically modeled as nonlinear programming problem.Finally, a deep investigation on mechanism of PSO is conducted. Not limited to the traditional velocity-position model, the GPSO model is proposed. Then, the limitation of information sharing mechanism of GPSO model is discussed. Based on this limitation, the improved information sharing mechanism for population based meta-heuristic algorithms is proposed. The experimental results on the application of solving permutation flow shop problem and open shop problem validate its effectiveness and superiority.