1. Photovoltaic power forecast using empirical models and artificial intelligence approaches for water pumping systems.
- Author
-
Ben Ammar, Rim, Ben Ammar, Mohsen, and Oualha, Abdelmajid
- Subjects
- *
WATER pumps , *ARTIFICIAL intelligence , *MAXIMUM power point trackers , *SOLAR energy , *FUZZY neural networks , *SOLAR radiation , *ALTERNATING current electric motors , *WEATHER forecasting - Abstract
The solar water pumping system is one of the brightest applications of solar energy for its environmental and economic advantages. It consists of a photovoltaic panel which converts solar energy into electrical energy to operate a DC or AC motor and a battery bank. The photovoltaic power fluctuation can affect the water pumping system performances. Thus, the photovoltaic power prediction is very important to ensure a balance between the produced energy and the pump requirements. The prediction of the generated power depends on solar irradiation and ambient temperature forecasting. The purpose of this study was to evaluate various methodologies for weather data estimation namely: the empirical models, the feed forward neural network and the adaptive neuro-fuzzy inference system. The simulation results show that the ANFIS model can be successfully used to forecast the photovoltaic power. The predicted energy was used for the solar water pumping management algorithm. • Temperature and solar irradiation forecasts are done for PV power estimation. • Ghouard's empirical model is the best for solar radiation forecasts on sunny days. • Capderou's empirical model is the best for solar radiation forecasts on cloudy days. • The ANFIS outperforms the empirical models and the FFNN in weather data prediction. • Based on the predicted PV power, water pumping management algorithm is done. [ABSTRACT FROM AUTHOR]
- Published
- 2020
- Full Text
- View/download PDF