Journal
ENERGY
Volume 115, Issue -, Pages 1672-1678Publisher
PERGAMON-ELSEVIER SCIENCE LTD
DOI: 10.1016/j.energy.2016.03.057
Keywords
Energy; Building; Heating load; Taguchi; Neural network
Categories
Ask authors/readers for more resources
Prediction of heating and cooling loads is necessary for building design and HVAC system operation, in order to reduce energy consumption. This study intended to develop a method for the prediction of the instantaneous building energy load, depending on various combinations of input parameters using a dynamic neural network model. The heating load was calculated for a typical apartment building in Seoul for a one-month period in winter using the Energy-Plus software. The data sets obtained were used to train neural network models. The input parameters included dry-bulb temperature, dew point temperature, direct normal radiation, diffuse horizontal radiation; and wind speed. The Taguchi method was applied to investigate the effect of the individual input parameters on the heating load. It was found that the outdoor temperature and wind speed are the most influential parameters, and that the dynamic model provides better results, as compared with the static model. Optimized system parameters, such as number of tapped delay lines and number of hidden neurons, were obtained for the present application. The results of this study show that Taguchi method can successfully reduce number of input parameters. Moreover dynamic neural network model can predict precisely instantaneous heating loads using a reduced number of inputs. (C) 2016 Elsevier Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available