Conference Paper
Proceedings of Building Simulation 2017: 15th Conference of IBPSA
The Potential of Artificial Neural Networks to Model Daylight Harvesting in Buildings Located in Different Climate Zones
The Potential of Artificial Neural Networks to Model Daylight Harvesting in Buildings Located in Different Climate Zones
Raphaela Walger da Fonseca1 , Fernando O. R. Pereira 1 , Konstantinos Papamichael 2 1 Environmental Comfort Laboratory, Federal University of Santa Catarina, Brazil 2 California Lighting Technology Center, UC Davis, USADOI: https://doi.org/10.26868/25222708.2017.570Abstract: Tools able to allow fast construction performance feedback in early design stages are essential to the design process. Artificial neural networks (ANNs) have high potential to metamodeling building energy performance considering daylighting harvesting. The aim of the present study was to assess the potential of ANNs to predict the energy performance of daylit buildings located in different climate. The ANNs were trained using results from parametric computational energy simulations, by taking into consideration the key daylighting parameters. Accordingly, the ANNs were able to predict the total energy consumption in different climates when cities from different hemispheres were taken into account. Pages: 2103 - 2112 Paper: