Prediction of daylight availability for visual comfort

Energy efficiency along with visual comfort have been given utmost importance in the present scenario. This has given rise to the control of daylight which allows the integration of daylight along with artificial light in any interior. The cost of energy can be reduced by combining lighting controls along with blind control. Thus research and development of intelligent prediction algorithms will help in development of automated daylight control systems. The predictive algorithms are considered due to the non-linear nature of daylight. This work elucidates and evaluates the daylight prediction strategy by using different techniques as time series prediction, nntool and nftool. Various daylight prediction models were developed in order to obtain the most efficient and accurate prediction method to achieve ideal visual comfort. In order to estimate illuminance in the interior due to daylight, the illuminance incident on the exterior of the window has to be predicted. Thus the value of illuminance and luminance has to be calculated from the irradiance and radiance data available. As a result exterior illuminance was evaluated using Perez model.