The Improved Of Particle Swarm Optimization And Its Application For Solving Nonlinear Problem

Optimization of industrial design and project implementation is frequently encountered problems, many problems can be attributed to optimization problem.It is often refer to the chemical production problems in engineering,technology,and economy,management and scientific research,and the solving method of the problem is paid close attention to people.The purpose is to find the conditions of the objective function's minimum or maximum.The conventional method is only applicable to that the practical solution domain is not very large and the problem is very clearly defined .But the real problems often have large search space, and these problems are not very clear definition, so the conventional method is either the accuracy or execution of time,is not always very satisfactory.Therefore, practical and effective optimization method is very necessary.The evolving commonly methods such as artificial neural network and the tabu search,simulated annealing, genetic algorithm, the ant colony algorithm and the particle swarm optimization algorithm in solving optimization problems shows unique advantages,they can get the complex problem optimal solution in a reasonable amount of time.The speed and the high precision is incomparable by conventional method.Particle Swarm Optimization(PSO) algorithm was developed by Kennedy and Eberhart in 1995, which has obvious ties with both evolutionary computation and swarm intelligence.The particle swarm optimization algorithm has the characteristics of evolutionary algorithm and swarm intelligence,its solving ability in the numerical optimization problem is unusual, so it becomes to one of the research hotspot of the evolutionary optimization algorithm.Article firstly analyzes the research background of the particle swarm optimization algorithm, and simply explains artificial life, artificial life calculation, swarm intelligence and the concept of intelligent algorithm,and then briefly introduces the particle swarm optimization algorithm based on several issues,including the origin,the original algorithm,the basic process,standard algorithm, etc. Later,Article simulates and analyses some parameters in the experiment, provides selection basis for the Practice parameters.Aim at hardness of initial parameters selection and easiness in local extremum,based on parametric optimization grouping particle swarm optimization algorithm has brought up. Aim at tardiness of convergence speed and accuracy, based on optimal direction of particle swarm optimization algorithm has put forward. Finally,the two algorithms are applied to nonlinear optimization problem and the problem of nonlinear equations, and compare to the standard particle swarm algorithm and other methods.Based on experimental analysis of the experimental data, article illustrates the effectiveness of these two kinds of algorithms.