Hybrid MPPT Technique Based on MPC and PSO for PV Systems Subject to Partial Shading

Photovoltaic (PV) systems are often subjected to partial shading, where a portion of PV array receives less radiation. When this occurs, local maximums appear on PV curve of the PV system. Algorithms based on local search such as P & O and INC are not able to reach maximum global power. For this type of application, are used global maximum power point tracking (GMPPT) algorithms. This work proposes a hybrid GMPPT applied to PV systems. It presents a hybrid approach using a global search technique and a local search technique. It uses the MPPT technique based on predictive control (MPC-MPPT) and the particle swarm optimization technique (PSO). The PSO algorithm is known to be an overall slow performance optimization method for the MPPT problem. To speed up your tracking capabilities, it is combined with the MPC-MPPT strategy, which is faster, but can get stuck at local maximums. The proposed method is tested in a SEPIC dc-dc converter connected to a partially PV panel subject to shading conditions. The results show the effectiveness of the proposed technique to accelerate the tracking ability of the PSO.