Experimental evaluation of global maximum power point techniques under partial shading conditions

Shading plays an important role in the operation of photovoltaic (PV) systems, especially when one of the cells or modules in the string is submitted to lower irradiance levels due to partial shading (PS). As a result of this condition, the cells or modules absorb the power generated by the unshaded parts, thus causing hot spots that can irreversibly damage the device. In order to mitigate this problem, bypass diodes (BDs) are often connected in parallel with the cells or modules to limit the reverse voltage and the power loss as a consequence. However, the occurrence of multiple peaks in the power versus voltage (P-V) curve is observed. Although conventional maximum power point tracking (MPPT) algorithms are successfully used when PV arrays operate under uniform irradiance conditions, they may fail to determine the global maximum power point (GMPP) if PS comes to occur. Considering that several MPPT algorithms for the operation under PS have been proposed during the last few years, this work presents a detailed analysis of some techniques dedicated to global maximum power tracking (GMPPT), aiming at defining which one presents the best performance. A dc-dc single-ended primary inductance converter (SEPIC) associated with microprocessor ATmega328P is employed to track the GMPP and assess the operation of each technique experimentally. The performance of the well-known perturb and observe (P&O) algorithm is also evaluated under PS condition for comparison purposes.