The Random Wander Ant Particle Swarm Optimization and Random Benchmarks

To solve the problem that the swarm was trapped by local optimization in searching process, the random wander ant Particle Swarm Optimization(called RWA-PSO) was proposed. The algorithm applied the mechanism of ant randomly wandering to find the food, and introduced it into the velocity updating process of particle. The probability that particle flied out the range of initialization increased. The local optimum can not trap the particles. The random benchmark and classical benchmark were applied in the numerical experiment to judge the performance of PSOs. The result showed that the RWAPSO had better searching results than the standard PSO and the typical CPSO. And it can solve the problem of premature convergence to a local optimum.