Abstract
This paper proposes an approach for finding the maximum power point (MPP) location of the photovoltaic (PV) array in the case of partial shading based on the shading pattern recognition. Shading pattern recognition is achieved by determining the necessary solar irradiation for each part of the PV array surface to recreate the existing power-voltage (P-V) curve of the PV array. This is done using the metaheuristic particle swarm optimization (PSO) method in which criterion function is based on comparing the measured powers of the PV array (fitting power points) and the powers calculated using the PV array model for the same voltages. Model of the PV array is created from the PV cell model considering the topology and the partial shading of the PV array surface. The effect of partial shading is included by reducing the photocurrent of the PV cells in the shaded area of the PV array. When the shading pattern is known the location (voltage) of the MPP is found using the P-V curve obtained from the PV array model. Results generated for three different shading patterns show great precision of the proposed approach in determining the MPP location if the PV cell is accurately modeled, and enough number of fitting power points are used.

This publication has 12 references indexed in Scilit: