Mixed Particle Swarm Optimization Algorithm with Multistage Disturbances

Abstract

In order to solve the problem that the particle swarm optimization algorithm is easy to fall into the local optimal value, a hybrid particle swarm optimization algorithm with multi-level perturbation is proposed. The algorithm combines the advantages of two classical improved particle swarm optimization algorithms, namely the standard with inertial parameters. Based on the particle swarm optimization algorithm and the particle swarm optimization algorithm with shrinkage factor, a multi-level perturbation mechanism is introduced. When the particle position is updated, the first-order perturbation is introduced to enhance the traversal ability of the particle to the solution space. In the case of “local optimal”, a second-order perturbation is introduced, which causes the optimization process to continue, thus getting rid of the local optimal value. Six test functions are used - Sphere function, Ackley function, Rastrigin function, Styblinski-Tang function, Duadric Function and Rosenbrock function to the proposed. The hybrid particle swarm optimization algorithm is used for simulation and comparison verification. The simulation results show that the proposed hybrid particle swarm optimization algorithm is better than the other two classic improvements in the simulation of the test function. Particle swarm optimization algorithm; in addition, when dealing with multimodal functions, the algorithm is not easily limited by local optimal values.