Prediction of temperature decreasing on a green roof by using artificial neural network

ERDEMİR D., Ayata T.

APPLIED THERMAL ENGINEERING, cilt.112, ss.1317-1325, 2017 (SCI-Expanded) identifier identifier

  • Yayın Türü: Makale / Tam Makale
  • Cilt numarası: 112
  • Basım Tarihi: 2017
  • Doi Numarası: 10.1016/j.applthermaleng.2016.10.145
  • Dergi Adı: APPLIED THERMAL ENGINEERING
  • Derginin Tarandığı İndeksler: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Sayfa Sayıları: ss.1317-1325
  • Anahtar Kelimeler: Green roof, Temperature decreasing, Thermal comfort, Artificial neural network, Roof cooling, PERFORMANCE, BUILDINGS
  • Erciyes Üniversitesi Adresli: Evet

Özet

This study presents an artificial neural network (ANN) model to predict temperature decreasing on a green roof. An ANN model has been created by MATLAB Neural Network Toolbox. The data for training, test and validation of the ANN model have been taken from nine different cities around the world. Meteorological data sets and temperature decreasing values on the green roof for these nine cities have been obtained from the study which is situated in literature. ANN model has indicated sufficient results with 0.3982% RMSE and 99.05% R-2. This ANN model has been used for estimating the temperature decreasing on the green roof for different cities from Turkey. At the end of present study, it is found that the temperature decreasing value for cities in Turkey shows same trend with world cities used in present study. While the minimum temperature decreasing is seen between 06:00am and 10:00am, the maximum temperature decreasing is seen at 06:00pm. The maximum temperature decreasing is determined in Kayseri with 33.28 degrees C. Results show that the green roof systems can be applied to all cities in Turkey. Because green roof supplies better thermal comfort and more energy saving. (C) 2016 Elsevier Ltd. All rights reserved.