Research And Application Of Particle Swarm Optimization Algorithm

Genetic algorithm is based on the population evolutionary mechanism and bio-genetic mechanism imitation in nature, individuals in the population has a probability operation of selection, mutation and crossover. Quantum theory reveals the law of the micro material world and provides a new representation method of nature for human, according to introduce the quantum theory into the study of intelligent algorithms, making particles has quantum behavior can overcome the shortcomings of the original algorithm. Nowdays, researching the quantum theory algorithm is very hot. Particle swarm optimization algorithm is a bionic optimization algorithm, it comes from birds looking for food in the nature, because of its few parameters, simple search process, and the good result of searching, academic circle has drawn closer attention to the study of the algorithm as soon as it is proposed, and the algorithm has been applied to the function optimization, neural network training, engineering applications, and many other areas. Nowdays, there are many improved algorithms, to some degree, these algorithms have been improved the performance, however, the convergence speed and how convergence to the global optimal solution, there are still lots of research space, so the study of the fusion algorithm of particle swarm optimization algorithm and genetic algorithm and the quantum theory, then enhancing the convergence speed and the effect of the optimization is very meaningful.With the market competition becoming more and more fierce, the mode of operation of traditional enterprise can not keep up the pace of change in the market, and base on the multiple independent enterprises temporary alliance of interest, that is, the virtual enterprise, by virtue of its flexibility and efficiency in the socio-economic, its status in business activities highlight and becomeing an important form of business organization. Thus, the key issues of the formation of virtual enterprise that how to choose the partners becomes particularly important. First of all, this paper introduces several intelligent optimization algorithms in detail, then studies the particle swarm optimization algorithm, and analyzes the advantages and disadvantages of the particle swarm algorithm; according to researching the quantum-behaved particle swarm optimization algorithm and the parameters of QPSO algorithm, this paper proposes a modified quantum-behaved particle swarms optimization algorithm and applies MQPSO to function optimization experiments. The results of these experiments show that the improved quantum-behaved particle swarm optimization algorithm has better stability and faster convergence; Based on the research of existing particle swarm optimization algorithm and its improved algorithms, combines with genetic algorithm, improving the defects of particle swarm optimization that easy to fall into local extremum and slow convergence, this paper proposes a genetic and particle swarm optimization hybrid algorithm, and applies the GPH algorithms to virtual enterprise partner selection simulation experiment to find the optimal combination of virtual enterprise partners, the experimental results show that it is an effective algorithm.