الفهرس | Only 14 pages are availabe for public view |
Abstract Renewable energy is essential for supplying the power needs of today’s generation. In reality, renewable source of energy is the only one that allows for power decentralization. The photovoltaic (PV) is one of a renewable energy. The output of a single PV Module is affected by environmental factors such as levels of irradiation and temperature. The main challenge of the PV system is tracking the maximum power point (MPP). Many techniques are used for maximum power point tracking (MPPT). Some of them are traditional and others are intelligent. The best results are found to be based on artificial intelligent (AI) techniques. In this thesis, a novel MPPT strategy that depends on two stages is presented. The first stage is to estimate values of temperature (T) and irradiance (G) using two different proposed artificial intelligent (AI) techniques which are artificial neural network (ANN) and fuzzy logic control (FLC). Voltage and current sensors are used instead of temperature sensor and pyranometer. This estimation achieves low cost and high accuracy. The second stage is to control the duty cycle (D) to track maximum power point using proposed ANN under variation of T, G, and load. The first stage for estimation T and G is investigated using seven case studies. Four of them are MATLAB simulation cases of proposed ANN and FLC. The other three cases are real case studied in Hurgada and Elsalam city. The second stage for tacking MPP is investigated using four case studies. Three of them are MATLAB simulation cases of proposed ANN. The fourth case is real studied in Hurgada. The results of tracking MPP are compared with three modified traditional methods. The output power of proposed method is higher, lower oscillation, higher efficiency and faster response. The proposed method is validated by experimental works. The experimental works are achieved in two different sites. The first experiment is in LAB, where the second is in external site (EAEAT academy). The experimental results are compared with simulation results. The results of experimental works verify the simulation results of the proposed methods with less error. |