Performance improvement of a photovoltaic system with a radial basis function network based on particle swarms optimization

This paper develops a RBF neural network based on particle swarm optimization (PSO) algorithm. The PSO algorithm is used to optimize the parameters of the RBFNN by introducing a new adaptive strategy of particle swarm optimizer to dynamically adjust the inertia weight factor ω and the new velocity vid(t+1) with a new μ coefficient. Therefore, the PSO algorithm has not only simplified the structure of RBF neural network, but also enhanced training speed and mapping accurate. The performance and effectiveness of the proposed method are evaluated by using function simulation and compared with RBF neural network. The results show that the optimized RBF neural network has significant advantages inspect of fast convergence speed, good generalization ability and not easy to yield minimal local results.