Simulation and experimental validation of fast adaptive particle swarm optimization strategy for photovoltaic global peak tracker under dynamic partial shading

The P-V characteristics of PV array has one peak under uniformly distributed irradiances. Whereas, there are many peaks in the P-V curve when the irradiance is not uniformly distributed over the PV array which is called partial shading conditions (PSCs). Due to its robustness in tracking the global peak (GP) of many applications, metaheuristic techniques are used as maximum power point tracker (MPPT) for the PV system under PSCs. Particle swarm optimization (PSO) has been used in this paper for this purpose. Three problems associated with the PSO have been solved in this paper using a novel fast adaptive PSO (APSO) strategy. The problem of long convergence time has been solved by updating starting values of the duty ratio of the DC-DC boost converter to be at the anticipated places of peaks. This modification reduces the convergence time and avoids the premature convergence. The problem of stored GP in the memory will prevent the PSO from capturing the current GP in case of it is lower than the stored one. This problem is solved in this paper by updating the memorized GP with the current maximum power when it is not changed for two successive iterations. The third problem of sudden change in PSCs is solved by using the updated values of duty ratio at anticipated peaks as initial values for particles. To the best of the authors' knowledge, these problems have not been discussed or solved before in the literature. A comparison to the state-of-the-art random initialization PSO strategy shows the superiority of the proposed APSO technique in terms of tracking speed and dynamic GP tracking. The results obtained from the simulation of this strategy proved its superiority in always tracking the GP under dynamic PSCs change.