Linear and Nonlinear Modeling for Solar Energy Prediction for Zone, Region and Global Areas

摘要

Solar energy data provides information on the sun's potential at a location during a specific time period. These data are very important for designing and sizing solar energy systems. Due to the high cost and installation difficulties in solar measurement, solar energy data are not always available. Therefore, there is a demand to develop alternative ways of predicting solar energy data. Here, we present linear and nonlinear models for global and diffuse solar radiation. The various models have been tested in different areas. These areas are classified as zone, region and global. The zone and region models were found to be accurate and could be used to predict solar radiation, which is a really interesting achievement. However, the global models had a high percent error. Three statistical values were used to evaluate the developed models, that is, root mean square error, the mean absolute percentage error (MAPE), and mean bias error. The results found the nonlinear models to be more accurate than the linear models - when calculating the solar energy in Malaysia using the nonlinear model, the MAPE was 6.4%; however, when using the linear models, the MAPE was 7.3%.