Simulation of a maximum power point tracking system for improved performance of photovoltaic systems

Abstract

In line with global trends, Kenya is increasingly turning to renewable sources of energy, namely Hydro, Solar, Wind and Geothermal to boost its capacity feed to the national grid. As such, Photovoltaic (PV) power generation has a key role to play. Compared to other renewablel sources of power (like wind and geothermal), PV power has the advantage of being installed in places that would otherwise have no other use such as roof tops. However, due to the equipment required, it still remains expensive to small scale dosmetic users. To counter this, delibarate e orts are being made to make them more competitive by increasing their e ciency. The major cause of energy losses in PV arrays has been recognized to occur when the PV arrays are operating at nonuniform conditions such as in a case where part of the modules are under shade while the rest are receiving the nominal solar radiation. Under partial shading conditions PV characteristics get more complex with multiple peaks. Because of multiple Maximum Power Points (MPPs), a considerable amount of available electrical energy may be lost when the module is operating at a local Maximum Power Point (MPP) with low power instead of the global MPP. Hence, it is necessary to understand the occurence of multiple MPPs under partial shading in order to extract maximum power. This thesis explores the e ect of partial shading on the power-voltage (P-V), powercurrent (P-I) and current-voltage (I-V) characteristic curves through models of the PV array created in MATLAB/SIMULINK software. It also sought to determine the e ectiveness of employing a conventional MPP technique, namely the Perturb and Observe (P&O) in tracking the MPP in the prescence of multiple peaks and compared results with that of a proposed two-tier maximum power point tracking technique. This technique aims to locate the global MPP on the P-I curve of the interconnected PV arrays by bypassing any local maximum that may trap the conventional MPPT schemes. The technique is split into two parts with the rst stage being used to nd a point that bypasses any local maximum and moves the operating point of the PV arrays near the global MPP. The second stage then nds the global MPP and sets the operating point of the PV arrays at this maximum. The system consists of two series PV array, A DC-DC converter, a load and the proposed technique. Operation of the model employing the proposed MPPT technique was veri ed by measurements of electrical characteristics of a PV array under partial shading. Simulation results show the existence of only one peak on the P-V and P-I characteristic curves and one step on the I-V characteristic curve under uniform insolation. However,under partial shading, there exists multiple peaks on the P-V and P-I characteristic curves and multiple steps in the I-V characteristic curve with power peaks being displaced from each other by a multiple of 8% of open circuit voltage. Simulation results also show that in the prescence of multiple peaks, the Perturb and Observe (P&O) can unexpectedly track the global peak, be trapped at a local peak for some time but continue to track the global peak or totally be unable to track the global peak. The results also showed that compared to the conventional P&O algorithm, the proposed MPPT algorithm provides improved performance in tracking the global MPP in the prescence of multiple peaks. Experimental results of model validation show that the model of the PV module used was reliable and accurate. Experimental implementation of the proposed MPPT algorithm indicates that a signi cant amount of additional energy ( an increase of about 130% of additional PV power) can be extracted from a PV module under partial shading when using the proposed technique as compared to when the P&O algorithm is used. This results in improved e ciency in the operation of the PV system, which is expected to increase cost savings in the long term. The results of this work can be applied to improve the operation of a partially shaded PV system by improving the maximum power point tracking of partially shaded PV arrays, resulting in the increase of power harvested from the PV system which is expected to increase cost savings in the long haul. Thus contributing to making PV-power generation a viable alternative to hydro-power generation.