| Summary: | An investigation into the design of a stand-alone photovoltaic water pumping system for supplying rural areas is presented. It includes a study of system components and their modelling. The PV water pumping system comprises a solar-cell-array, DC-DC buck chopper and permanent-magnet DC motor driving a centrifugal pump. The thesis focuses on increasing energy extraction by improving maximum power point tracking (MPPT). From different MPPT techniques previously proposed, the perturb and observe (P&O) technique is developed because of its ease of implementation and low implementation cost. A modified variable step-size P&O MPPT algorithm is investigated which uses fuzzy logic to automatically adjust step-size to better track maximum power point. Two other MPPT methods are investigated: a new artificial neural network (ANN) method and fuzzy logic (FL) based method. These use PV source output power and the speed of the DC pump motor as input variables. Both generate pulse width modulation (PWM) control signals to continually adjust the buck converter to maximize power from the PV array, and thus motor speed and the water discharge rate of a centrifugal pump. System elements are individually modelled in MATLAB/SIMULINK and then connected to assess performance under different PV irradiation levels. First, the MP&O MPPT technique is compared with the conventional P&O MPPT algorithm. The results show that the MP&O MPPT has faster dynamic response and eliminates oscillations around the MPP under steady-state conditions. The three proposed MPPT methods are implemented in the simulated PV water pumping system and compared. The results confirm that the new methods have improved energy extraction and dynamic tracking compared with simpler methods.
|
|---|
