Comparison between conventional methods and GA approach for maximum power point tracking of shaded solar PV generators

Fuzzy-predictive direct power control implementation of a grid connected photovoltaic system, associated with an active power filter

“…In addition, it presents noticeable tracking skills in case of large and sudden irradiance variations. In case of partial shading effect, where the incident irradiance exhibits multiple plateaus, heuristic global optimization methods, such as particle swarm optimization (PSO) [12] and Genetic algorithm (GA) [13], are considered the more suitable, since a global extremum is always obtained through a random search pattern. As described in [12], the well-known 'distributed MPPT architecture' is proposed to improve the energy production with respect to a centralized MPPT technique, and offers additional features in term of protection.…”

Sustainability and process control: A survey and perspective

“…Periodic scanning of the current-voltage curve may be used to identify the global maxima [294,302], but a significant amount of available power may be lost depending on the frequency of scanning. Advanced techniques can be applied to search for the global maximum power point including multi-stage hill climbing algorithms [303,304], particle swarm optimization [305,306], genetic algorithm [307], fuzzy logic [308,309], and artificial neural networks [308]. These techniques improve the amount of PV power harvested in partial shading conditions, but are more complicated to implement and take more effort to properly tune/train.…”

Modified Particle Swarm Optimization technique based Maximum Power Point Tracking for uniform and under partial shading condition

“…In the recent past, bio-inspired algorithms like GA, PSO and ACO have drawn considerable researcher's attention for MPPT application; since they ensure sufficient class of accuracy while dealing with non-linear, non-differentiable and stochastic optimization problems without involving excessive mathematical computations [32,[36][37][38]. Further, these methods offer various advantages such as computational simplicity, easy implementation and faster response.…”