Logarithmic PSO-Based Global/Local Maximum Power Point Tracker for Partially Shaded Photovoltaic Systems

It is well known that under partial shading conditions, photovoltaic generator (PVG) power-voltage characteristic presents over one maximum due to the use of bypass diodes. Classical hill climbed methods cannot found the global maximum and often converges to a local maximum. Among numerous optimization methods used to track the global maximum, particle swarm optimization (PSO) has been widely used due to its simplicity and efficiency. However, this method suffers from a large amount of power ripple during the search process and slow convergence speed. Several works tried to solve this problem by reducing the search window that is often a complex procedure and can lead to a lack of generality. This article presents a logarithmic PSO method, and this method has been used to design a maximum power point tracker (MPPT) combining global and local MPPTs. It reduces power oscillations during the search process and accelerates the convergence without search window reduction, and only one parameter has to be tuned, which facilitates the design. During steady state, the swarm is reduced to one particle, which slightly perturbs the PVG to detect small and slow local variations of the maximum power point. Simulations and experimental results show the effectiveness of the proposed method.