Performance Comparison of Root-finding Algorithms for Solar Photovoltaic Emulator

Abstract

Solar photovoltaic (PV) emulator is a non-linear power supply which emulates the characteristics of a solar PV module. Solar PV emulators are widely used in laboratories instead of actual PV modules for testing various solar PV-based systems. Using a suitable mathematical model of solar PV module, the PV emulators are expected to produce similar voltage-current characteristics of the commercially available solar PV modules. An iterative root-finding algorithm is used to compute the value of the output current of solar PV module using solar irradiance, temperature and cell voltage as inputs. This root-finding method always causes a delay in computing the actual value of the current through the iteration process. Hence, this delay introduces a sluggish response in the emulator particularly when it is operated in the boundary region of the solar PV characteristics. In this work, a comparison is made between three different root-finding algorithms such as Newton-Raphson, binary search and secant method. The performance of the above three methods are verified through simulations. In PV emulators, conventional direct referencing method and current-resistance method are used. In both the methods, all the three root-finding algorithms are applied and the best method is found. For conventional direct referencing model, the Newton-Raphson method is converging faster out of three root-finding algorithms, whereas binary search is fast converging in case of current-resistance model.