A Fast RBFM Based MPPT Method for Application of Triple-Junction Solar Cell

This paper proposes an efficient maximum power point tracking (MPPT) technique for the photovoltaic (PV) system with a particle swarm optimization based radial basis function network (PSO-RBFN) algorithm. In PSO-RBFN, the weights of RBFN is trained in the first step with PSO, which is well recognized as an efficient optimization algorithm for continuous problems, and then the RBFN weights continue to be trained until the weights are optimum for the whole network, which leads to this technique improves the training speed and accuracy. Combined with a PV model based on the triple-junction solar cell formed by masked block in Matlab/Simulink, the presented PSO-RBFN MPPT is proposed in the power system. A comparison of the PSO-RBFN with radial basis function network (RBFN) MPPT, back propagation neural network (BPNN) MPPT and perturb and observe(P&O) MPPT under the same condition and the same model is made. The results show that the output power and energy of the proposed technique are higher than that of P&O MPPT technique and the performance of training is better than that of RBFN MPPT technique and BPNN MPPT technique.