Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Architectural Institute of Japan  

This study utilizes CFD models to assess the indoor thermal environment and energy loads of grocery stores of varying sizes. The models simulate indoor conditions under typical summer and winter climates, examining how changes to air conditioning settings and ventilation rates affect energy consumption and thermal comfort. The study also investigates the effects of ceiling height and the presence of a windbreak room. Results indicate that larger stores are less influenced by outdoor air infiltration, whereas smaller stores exhibit significant vertical temperature variations. Additionally, ceiling height and the presence of a windbreak room further impact indoor temperatures and cooling loads.

DOI： 10.3130/aije.90.360

researchmap

Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.buildenv.2025.113616

researchmap

Authorship：Lead author,　Corresponding author   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.buildenv.2025.113246

researchmap

Publishing type：Part of collection (book)   Publisher：Springer Nature Singapore  

DOI： 10.1007/978-981-97-8317-5_4

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1016/j.clrc.2024.100244

researchmap

Language：Japanese   Publisher：関東学院大学大沢記念建築設備工学研究所  

To advance the intelligent management of air conditioning systems, there is a growing demand for indoor airflow prediction methods that reduce computation time compared to traditional numerical fluid dynamics (CFD) simulations. Recently, artificial neural networks (ANNs) have garnered significant attention in the industry due to their ability to learn from limited CFD data and efficiently compute numerous scenarios under varying air conditioning conditions. However, ANNs often replicate data patterns without comprehending the underlying physical phenomena, leading to potential discrepancies between predictions and actual physical realities. Physics-informed deep learning addresses this issue by integrating physical equations with data, enabling the model to produce more reliable and physically consistent predictions. This study introduces physics-informed deep learning and explores its potential applications in the architectural domain.

CiNii Books

researchmap

Other Link： http://id.ndl.go.jp/bib/034182330

Language：Japanese   Publisher：The Society of Heating,Air-Conditioning&Sanitary Engineers of Japan  

DOI： 10.18948/shasetaikai.2024.03.0_37

researchmap

Language：Japanese   Publisher：Institute of Industrial Science The University of Tokyo  

This study aims to develop a fast and accurate indoor airflow prediction method using deep neural network (DNN). Here, the effect of training data selection on transient prediction is investigated. Five data selection scenarios were considered to find out a method selecting training data, which is benefit to training DNN efficiently, from two aspects: (1) time selected for each case, (2) selection of cases with different setting conditions. As a result, it is better to take time points into consideration as many as possible for each case. At the same time, it is better to take cases in the interior of condition space (a coordinate space formed by setting conditions of condition change experiment case) as training cases.

DOI： 10.11188/seisankenkyu.74.53

CiNii Article

CiNii Books

researchmap

Language：Japanese   Publisher：The Society of Heating,Air-Conditioning&Sanitary Engineers of Japan  

DOI： 10.18948/shasetaikai.2021.5.0_133

researchmap

Language：Japanese   Publisher：日本建築学会  

CiNii Article

researchmap

Event date： 2024.10.11

Language：Chinese   Presentation type：Oral presentation (invited, special)  

researchmap

researchmap