Reducing Computational Costs and Mitigating Measurement Errors in Parameters’ Estimation for the Single Diode PV Model

Abstract Efficient modeling of PV systems is one of the key characteristics that enable this sector. The quality and robustness of the electrical model are strongly dependent on the accurate estimation of the physical characteristic parameters of the PV cell/module. However, considering the complexity of such a model, the computational cost is a significant concern. This paper presents effective techniques to minimize the computational cost by targeting multiple key aspects. One concept lies in the problem formulation phase, and the second one aims to optimize the number of the measured I-V points per curve, for which the physical parameters are to be extracted. The presented study has proved that using 20 I-V measured points per curve would reduce the computational cost for extracting the parameters by several folds compared to the typical situation in a measurement system that records 450 points per curve. Also, the suggested method shows its applicability in minimizing the error introduced by measurement systems.