COMPARATIVE ANALYSIS OF MPPT ALGORITHMS FOR PHOTOVOLTAIC SYSTEMS

Abstract

One of the most viable renewable energy sources is photovoltaic (PV) energy which serves as an alternative to fossil energy as it is considered less polluted. The PV systems must be operating with high efficiency. However, PV panels have a non-linear voltage-current characteristic, which depends on environmental factors such as solar irradiation and temperature and gives very low efficiency. Therefore, it becomes crucial to harvest the maximum power from the PV panels. Thus, they have to operate at their maximum power point (MPP) despite the inevitable changes in temperature and solar irradiation. The objective of this thesis is to study and analyze the MPPT techniques. The most common and well known perturb and observe (P&O), and incremental conductance (In Cond) methods were focused on in this thesis, as these algorithms were found easy to implement, low-cost techniques, and suitable for large size and medium-size photovoltaic applications and were compared with intelligence-based techniques that are Particle Swarm Optimization (PSO), Fuzzy logic and Artificial Neural Network (ANN). All these five techniques were implemented in the MATLAB Simulink environment and their performance was analyzed for variable irradiances at standard temperature. The proposed MPPT schemes are implemented in the control circuit of the DC-DC boost converter. Results show an improvement in the tracking of ANN based controller compared to all other controllers. The proposed algorithms minimize the oscillations around MPP, and the power is converging faster compared with the response done by conventional algorithms.