Comparison of regression model and artificial neural network model for energy benchmarking of accommodation buildings in kanto, Japan

Haitham Alkhalaf, Wanglin Yan

研究成果: Conference article

抜粋

Energy performance of residential and non-residential buildings is a vital topic because of fast urbanization in the world. The accommodation buildings are considered as high energy intensive comparing to other commercial building' categories. In addition, it has an important contribution in tourism industry. Therefore, variety of models and plans have been applied to reduce the energy consumption of accommodation buildings. This research depends on database of energy consumption of commercial buildings in Japan as main source of data. Data-base for Energy Consumption of Commercial building (DECC) is a national survey, it is disclosed by Japan Sustainable Building Consortium (JSBC). Base on DECC, a benchmark system is developed by applying regression and Artificial Neural Network (ANN) methods to assess the energy performance of accommodation building in Kanto region-Japan. The study investigate the primary energy model of selected samples according to consumption' trends of electricity, gas and clean water. The developed benchmarks by ANN and regression models were compared to ensure a robust benchmark system as a powerful tool for energy performance' assessment. This study points out the necessity to benchmark the energy performance of accommodation buildings and other categories in Japan. In addition, it is important to consider other variables that affect energy use of buildings.

元の言語English
ページ(範囲)71-82
ページ数12
ジャーナルWIT Transactions on Ecology and the Environment
224
発行部数1
DOI
出版物ステータスPublished - 2017 9 20
イベント7th International conference on Energy and Sustainability, ESUS 2017 - Seville, Spain
継続期間: 2017 9 202017 9 20

ASJC Scopus subject areas

  • Environmental Science(all)

フィンガープリント Comparison of regression model and artificial neural network model for energy benchmarking of accommodation buildings in kanto, Japan' の研究トピックを掘り下げます。これらはともに一意のフィンガープリントを構成します。

  • これを引用