Akademik Veri Yönetim Sistemi
Tezin Türü: Yüksek Lisans
Tezin Yürütüldüğü Kurum: Çukurova Üniversitesi, Fen Bilimleri Enstitüsü, Elektrik-Elektronik Mühendisliği, Türkiye
Tezin Onay Tarihi: 2023
Tezin Dili: İngilizce
Öğrenci: BATUHAN GÜZEY
Danışman: Ahmet Teke
Özet:
Renewable energy sources (RES) are critical to meet energy demand in the changing and
developing world. Among these resources, one of the most invested areas is solar energy.
Photovoltaic (PV) panels are devices that convert radiation from the sun into electricity. Temperature
and radiation parameters are used to generate electricity. Therefore, the produced power varies under
bad weather conditions. Maximum power point tracking (MPPT) algorithms are used with the PV
system to prevent power variability. In addition, battery-supported designs are utilized to ensure
power continuity at night and keep the system running. The aim of this study is to design a hybrid
PV system with grid connected and battery-supported. Moreover, in this thesis has been selected the
most efficient method among the implemented tracking algorithms. Perturb&Observe (P&O),
Incremental Conductance (INC), Artificial Neural Network (ANN) methods are applied respectively.
The ranges in which these three tracking algorithms perform the best results are combined and the
hybrid MPPT is designed. The system outcomes is tested at 1000-800-600 W/m2
irradiance levels in
MATLAB/SIMULINK program. Moreover, the battery energy storage system (BESS) has been
developed to ensure power continuity. Since the powers of PV and BESS are different from each
other, it has been determined that total harmonic distortion (THD) increases when the system is in
night mode. To solve this problem and improve power quality, two mode LCL filter design has been
performed. The system charges the battery while in daytime mode, and the daytime mode LCL filter
is activated. When the system is in night mode, the battery is discharged and the night mode LCL
filter is activated. Thanks to these designs, the THD value has been significantly reduced and the
most efficient MPPT algorithm has been proposed. When the results are compared, the THD value
obtained at 4.8% level while using single mode LCL filter. This value is decreased to 3.2% in two
mode LCL filter application. In addition, according to the test results, when ANN MPPT is operated,
97.5% efficiency is observed at low irradiance. In the circuit operated with high efficiency, the DC
link voltage has been become more uniform, smoothness.