Modeling and simulation of model predictive control based maximum power point tracking for photovoltaic panels

With the increase in the world population, industrialization and the development of technology, the amount of energy consumed is increasing day by day. Fossil sources account for more than 79% of the energy produced. Therefore, utilizing clean energy sources is an important solution. Solar energy is the most widely used clean energy source. The clean, inexhaustible and ease of use of solar energy makes solar energy more advantageous among other renewable energy sources. Direct electricity generation from solar energy occurs with photovoltaic panels. Photovoltaic literally means the production of electricity from a photon of light. Photovoltaic solar cells, on which photons fall, are systems that directly convert solar energy into DC electrical energy with the help of semiconductor materials. Although semiconductor technology has developed recently, the efficiency of solar panels is low. There are many factors that affect the efficiency of photovoltaic panels. These; panel tilt angle, shading, dusting, solar radiation intensity, temperature, cabling etc. are losses. In this thesis, a simulation study was carried out in MATLAB/Simulink program by using the Perturb &Observe Algorithm, one of the Maximum Power Point Tracking methods, in order to maximize the output power of a PV panel. A buck converter is used to reduce the sinusoidal DC voltage at the panel input and to ensure stable use. Model Predictive Control is used to control the control of all possible switching states of this power converter and to predict and optimize the future behavior of input-output voltages and currents. The MPC transfers the values to the DC-DC step-down converter to obtain the maximum power. Coil current is estimated rather than measured. This eliminates the use of extra current sensors and reduces the system cost. In the study, the changes of the PV panel according to different temperature and solar radiation values were analyzed with MPPT. The effects of the panel current, voltage and power curves obtained as a result of these analyzes on the performance of the PV panel were evaluated. It has been seen that the simulation results of the PV panel with step-down and step-up converters are around the Maximum Power Point. In the study, it was seen that the D&G algorithm gave consistent results at maximum power point levels. In addition, similar to other studies, it has been observed that the D&G algorithm has oscillations around the MPP. In addition, when we look at the transient status of the MPC, it is understood that it reaches the MGN levels quickly and is a good controller.