S. S. Mortazavi - Electrical Engineering Department, Faculty of Engineering, Shahid Chamran University, Ahwaz, Iran
P. Mohamadi - Department of Business Management, Faculty of Economics and social scienceShahid Chamran University, Ahwaz, Iran
A Rostami - Electrical Engineering Department, Faculty of Engineering,Shahid Chamran University, Ahwaz, Iran

Due to the growing demand on renewable energy,photovoltaic (PV) generation systems have increased considerablyin recent years. However, the power output of PV systems isaffected by different weather conditions. Accurate forecasting ofPV power output is important for power system reliability and management. Due to various seasonal, hourly and daily changes in climate, it is relativelydifficult to find a suitable analytic model for predicting thepower outputofGrid- Connected Photovoltaic (PV) plants.In this article, the ordinary least square regression (OLS) and the artificialneural network configuration is used for estimating the power produced by a20kWp PV plant installed at Ahwaz, Iran.We use irradiance, temperature, wind speed and dust amount as independent variables which influencepower output of PV plant. Results show that these factors are significant and while two methods have suitable performance, performance of neural network is better than ordinary least square regression. Also, correlation between power output and dust as localfactor is 7 percent.We use Eviews and MATLAB software to analyze data.

Photovoltaic, PV plant, Artificial neural network (ANN), Ordinary least square regression (OLS)