ABSTRACT For design and development of solar-based photovoltaic (PV) system, accuracy in obtaining data of solar energy has been a prime concern. Forecasting of global solar energy in clear or sunny weather conditions are easily calculated; however, the challenge lies in estimating under the impact of foggy and cloudy conditions. Therefore, this research investigates and presents the Artificial Neural Network (ANN) models for global solar energy forecasting for distinct sky-conditions such as sunny, hazy, partly, and fully cloudy conditions. This research further applied for distinct climatic conditions namely composite, warm & humid, moderate, hot & dry, and cold & cloudy climate zone using meteorological parameters. Further, PV power forecasting has been essential for planning the operating capability of energy systems and its reliability. The present work explores the implementation of proposed model for short-term photovoltaic power forecast in solar energy systems. Lastly, comparative analysis of the developed model with other neural network models are carried out using statistical evaluation indices. It has been observed from the overall analysis that by using the Radial-Basis-Function-Neural-Network (RBFNN) model, a significant reduction in error has been noticed as compared to other methods. Further, it is noted that the sunny model provides improved performance for hot & dry climate zone; hazy model gives favorable output for composite climate zone; partly cloudy model is favorable for warm & humid climate zone; and lastly, fully cloudy model provides favorable results for cold & cloudy climate zone. The model has been implemented for broader application series and it has been observed that for short-term photovoltaic power forecasting under composite climatic zone, hazy model emerges out to provide better performance with Mean Absolute Percentage Error (MAPE) of 0.0019% followed by sunny, partly, and fully cloudy model with MAPE 0.024%, 0.054%, and 0.109% respectively. The models presented in this research will be useful for scholars to design and engineer the solar energy systems.