Research On Modified Particle Swarm Optimization And Its Engineering Application

In the actual production and life,the optimization problems that people faced are increasingly complex and diverse,and traditional optimization algorithms cannot solve these problems.Inspired by the evolutionary laws of survival of the fittest in the natural world,and with the rapid development of computer technology and artificial intelligence,people began to use computers to simulate natural biological evolution processes and intelligent movement behaviors,and constructed various intelligent algorithm models to solve optimization problems,intelligent optimization algorithms came into being.Particle Swarm Optimization(PSO)is a global optimization algorithm that simulates the foraging behavior of flock birds.It has many advantages such as fewer parameters,simple operation and good stability.It attracts people's attention and applies it to different fields successfully.However,the particle swarm optimization algorithm also has the following defects: firstly,it is prone to "premature" phenomenon when solving complex optimization problems;secondly,most of the engineering optimization problems in actual production are constraint optimization problems,and the particle swarm optimization algorithm is an unconstrained optimization algorithm,it is necessary to use constraint processing to solve constraint problems before using it to optimize.Therefore,avoiding the algorithm "premature" and the processing of the constraint conditions are two important directions for studying the particle swarm algorithm and it is also the main problem to be solved in this paper.The main research content is as follows:(1)In order to solve the problem that the particle swarm optimization algorithm is likely to fall into local optimum,an improved particle swarm algorithm based on dynamic segmentation and neighborhood reverse learning is proposed.By setting up a dynamic segmentation mechanism,the algorithm divides the particles in the population into three grades,then employs different perturbation strategies for the particles in different grades,so that the particles maintain the evolution to the global optimal direction while the diversity of the population is enhanced.Furthermore,it adopts the method of particle intelligent updating to promote the search ability of particles,and introduces the dynamic neighborhood reverse point enabling a global search to improve the particle searching speed.The preliminary results show that the proposed algorithm has better convergence and stability than several other kinds of optimization algorithms.(2)Particle Swarm Optimization is an unconstrained global optimization algorithm.Considering that this algorithm is easy to fall into local optimum and lacks the drawbacks of constraint processing techniques,this paper proposes a modified particle swarm optimization algorithm for engineering constrained optimization problems.The algorithm first constructs an evaluation criterion of particle according to the fitness of the particle and the constraint violation degree.Then use the teaching mechanism to guide the population to evolve in a feasible direction,and make the dynamic variation of inertia weight and learning factors adaptive to balance the exploration and mining capabilities of the algorithm.Finally,a particle reset strategy is adopted to avoid the stagnation of the algorithm update.Simulation experiments were carried out on three typical engineering optimization designs.The results show that the algorithm can effectively solve different constrained optimization problems.