Comparative Estimation of Global Solar Radiation over Two Nigerian Cities, Using Artificial Neural Network and Empirical Models

The estimation of solar radiation intensity has been a focus many researchers due to the cost setting up its actual measurements. While them employed empirical models, this study utilizes artificial neural network for analysis and global over two Nigerian cities. model developed using sunshine hours, temperatures relative humidity were compared with existing models. Model performance indicators comparing measured data computed derived selected same number input meteorological parameters showed that ANN having average values RMSE, MBE, MPE 0.0744 MJm-2day-1, -0.0020 -0.0043%, respectively, performed slightly better. When different used, gave following error 0.0394MJm-2day-1, -0.0023MJm-2day-1 -0.0144% respectively. Also, in result Abuja, parameters, best is 0.1301MJm-2day-1, 0.0053MJm-2day-1 0.0441% Hence, models are versatile predicting locations climatic zones as studied study, where direct measurements scarce widely separated but there availability commonly such duration, minimum temperature, maximum temperature humidity.