In the past few decades, the power demand is increasing rapidly due to urbanization and industrialization. Day by day the conventional fossil fuels used for the generation of electricity is getting depleted. To meet this increase in load demand, generation of electricity through renewable energy sources like solar, wind, etc., is gaining a lot of importance in these days. Apart from central and state generation companies, many customers started generating solar power to meet their own load demand and excess power is exported to the grid. Since the solar power generation depends on atmospheric conditions and the generation is intermittent in nature and often accurate prediction becomes difficult. Also, it becomes difficult to power system operators to anticipate the solar power generation due to up and downs in the variable renewable energy generation, which is posing lots of challenges in integrating the solar power to the grid. Solar Power generation forecasting plays a vital role in reducing stability issues, helps power system operators to plan and schedule the generation of renewable energy, conventional energy generation. In this paper a model is developed using artificial neural network for forecasting the solar energy generation and an attempt is made to provide better accuracy compared to existing conventional forecasting models. [ABSTRACT FROM AUTHOR]