Multi-angle Improved Particle Swarm Optimization And NP Problem Application

With the development of science and technology,Traditional optimization methods often have some limitations when solving optimization problem.The intelligent optimization algorithm provides a new idea to the optimization problem.Particle swarm optimization(PSO)is a stochastic intelligent optimization algorithm based on the simulation of natural populations.Compared with traditional optimization algorithms,PSO has many advantages,such as simplicity of flow,few control parameters and easy implementation.However,since the theoretical basis of PSO is still far from mature,the problems with the premature convergence remain for exploiting,leading further to more spaces to improve while been applied to practical engineering.Based on studying PSO theory and practicing,this article mainly from the particle algorithm implementation process,particle parameters update such as inertia weight and learning factor,and some strategies such as opposition-based learning,disturbance,Gaussian mutation,Cauchy mutation,and studying particle swarm algorithm combined with other intelligent algorithm.Simulation experiments had been down from the parameters improvement,strategies improvement,and the application of particle swarm optimization.Compared with some well-known improved algorithms,the improved algorithm has advantages.The main innovations can be summarized as follows: Firstly,in the aspect of particle swarm optimization algorithm parameter improvement,the elite opposition-based particle swarm optimization based on disturbances was proposed.By enhancing the guiding effect of the elite particle,changing the weights in a non-linear way to balance the particles search and using a disturbance approach to enhance the ability of local exploration.Secondly,in the aspect of particle swarm optimization algorithm strategy improvement,the particle swarm optimization algorithm based on multi-strategy synergy was proposed.The strategy used is determined by the probability threshold.When the threshold is smaller than the threshold,the reverse learning strategy is adopted.When the threshold is greater than the threshold,the Gaussian mutation strategy is adopted to enhance the diversity of the population.While using the Cauchy distribution proportional parameter linear change strategy to guide the particles to the optimal solution space,Synergistic use of the advantages of each strategy to improve algorithm performance.Thirdly,in the aspect of particle swarm optimization algorithm specific application,the improved ant colony-particles swarm hybrid algorithm for solving TSP was proposed.By using the greedy algorithm initial particle position,and using adaptive crossover mutation strategy with the global optimal particle to avoid particles into the local optimal.