Artificial Neural Network based MPPT Algorithm with Boost Converter topology for Stand-Alone PV System

Abstract

The increasing energy need in parallel with the technology development and the depletion of the resources have increased the importance of alternative energy resources. Solar energy systems are frequently preferred due to their advantages such as not having moving parts, being reliable and working without noise. Production of electricity from solar energy is obtained by serial or parallel connection of photovoltaic (PV) panels, depending on the desired voltage and current values. DC-DC converters are used to convert the energy obtained from the PV panels to the desired grid values. Maximum power point tracking (MPPT) algorithms are used in order to obtain the highest possible efficiency from the PV panels. MPPT algorithms control the duty period (D) ratio of DC-DC converters and obtain maximum energy. In this study, an Artificial Neural Network (ANN) based MPPT algorithm is proposed. Firstly, the temperature and irradiance data at the PV panel input are trained using the Levenberg-Marquardt algorithm. As a result, a reference voltage is generated and MPPT is realized by comparing it with the voltage produced by the PV panel. In order to evaluate the performance of the proposed algorithm, it is compared with the traditional MPPT methods such as Perturb & Observe (P&O) and Incremental Conductance (INC). As a result of the simulation studies, it has been observed that ANN based MPPT is more successful than P&O and INC algorithms for several irradiance and temperature conditions.