Increasing the power obtained in wind turbines with maximum power point tracking methods

The transformation of countries from agricultural and animal husbandry societies to industrial societies or the increase in this ratio in favor of industry, the increase in the human population, the growth of cities, the growth of people's comfort areas and the differentiation of their needs lead to an increase in the need for energy. On the other hand, the fact that fossil energy production resources are ultimately limited and decreasing day by day makes it compulsory for countries to make new energy plans, to develop and use their renewable energy resources more efficiently, to increase the proportion of renewable energy in existing energy resources and to produce energy policies in the direction of diversification. Increasing the rate of use of renewable energy resources compared to fossil resources will also be effective in reducing carbon emissions and preventing or slowing down the deterioration of the balance of nature. While the human population is increasing, it will contribute to the healthy aging of people. Wind energy, which is the subject of this thesis, is one of the leading renewable energy sources. The basic parameter of wind power plants is the wind. The speed and direction of the wind are stochastic and can change daily, hourly or even instantaneously. In this study, our main objective is to apply variable wind speed scenarios to the wind power plant model that we have created using MATLAB/Simulink to ensure that the power value that the system can generate is always at the maximum power point. For this purpose, we used four different maximum power point tracking methods, namely incremental conductivity, perturb and observe, fuzzy logic and artificial neural network, and also compared the average power values obtained from the plant output without using any method. These five different cases were simulated with three different wind scenarios. Thus, in terms of both instantaneous power and average power values, the highest values obtained among the methods used were obtained from the artificial neural network method trained with the fuzzy logic method, followed by the fuzzy logic method. As a result, it is shown that the value obtained from the wind turbine output reaches the maximum value in terms of instantaneous and average power values using the methods examined in the thesis, which is a good result for the purpose of the thesis.