Improved cooperative artificial neural network-particle swarm optimization approach for solar photovoltaic systems using maximum power point tracking

Photovoltaic (PV) energy represents one of the most important renewable energies, but its disadvantage resides in its maximum power point, which varies according to meteorological changes that make the efficiency low. Intelligent techniques, using the maximum power point tracking (MPPT) method, can achieve an efficient real-time tracking of this point in order to ensure optimal functioning of the system. The output power of the PV system is removed from solar irradiation and cell temperature of the PV panel type SOLON 55W. Therefore, it is essential to harvest the generated power of the PV system and optimally exploit the collected solar energy. For this objective, this work treats on a new artificial neural network-particle swarm optimization approach (ANN-PSO). The ANN is used to predict the solar irradiation level and cell temperature followed by PSO to optimize the power generation and optimally track the solar power of the PV panel type SOLON 55W based on various operation conditions under changes in environmental conditions. The simulation results of the proposed approach give a minimum error with a relevant efficiency, that is, the power provided by ANN-PSO approach is optimal and closer to the PV power. Consequently, this novel approach ANN-PSO shows its major capability to extract the optimal power with excellent efficiency up of 97%. For this objective, this work treats a new hybrid ANN-PSO approach.