Time Series Analysis, Forecasting and Control
Time Series Analysis, Forecasting and Control
Artificial Intelligence technique for modelling and forecasting of solar radiation data: a review
International Journal of Artificial Intelligence and Soft Computing
| Hi-index | 0.00 |
The development of the models for global solar radiation in Kuwait is based on removing the annual periodicity and seasonal variation. The first methodology used here is the change in regime technique that relies on dividing the observations into two states, one for clear days and other for non-clear days depending on the duration of the sunshine. Then, for each state, the trend component is expressed as a Fourier series and the stochastic component as an ARMA model. The second methodology used is the transfer function model which depends on two same behavior and highly correlated series such as sunshine duration and daily global radiation. The sunshine duration is considered as an independent input series whereas the global radiation series as the dependent output series. A comparison is made between the two proposed models based on Akaike's Information Criterion (AIC), the Bayesian Information Criterion (BIC), and the coefficient of determination (R^2).