Construction of digital twin that fuses AI and physical model to optimize building energy system

2023 Fiscal Year Final Research Report

• Project/Area Number: 20H00273
• Research Category: Grant-in-Aid for Scientific Research (A)
• Allocation Type: Single-year Grants
• Section: 一般
• Review Section: Medium-sized Section 23:Architecture, building engineering, and related fields
• Research Institution: The University of Tokyo
• Principal Investigator: Ooka Ryozo 東京大学, 生産技術研究所, 教授 (90251470)
• Co-Investigator(Kenkyū-buntansha): 長井 達夫 東京理科大学, 工学部建築学科, 教授 (00316001) ; 池田 伸太郎 東京工業大学, 環境・社会理工学院, 准教授 (00843525) ; 菊本 英紀 東京大学, 生産技術研究所, 准教授 (80708082) ; 崔 元準 東京大学, 生産技術研究所, 助教 (30817458)
• Project Period (FY): 2020-04-01 – 2023-03-31
• Keywords: デジタルツイン / 人工ニューラルネットワーク / 最適化システム / メタヒューリスティクス

Outline of Final Research Achievements

In this research, in order to build a digital twin that mimics the energy system in a real building, the following three items have been investigated: (1) development of modeling methods for building components, (2) development of fully automatic control methods based on optimization tools, and (3) development of an integrated platform for the above, i.e. a digital twin. Regarding the modeling method (1), we developed a method based on artificial neural networks (ANN) with the aim of real-time prediction. Regarding (2), we investigated methods based on metaheuristics and reinforcement learning, and confirmed their characteristics. As to (3), the development items (1) and (2) were combined, and a digital twin of the building energy system of an actual building was constructed to create an integrated platform that connects virtuality and reality.

Free Research Field

建築環境工学

Academic Significance and Societal Importance of the Research Achievements

人工知能の特性とその建築設備応用可能性の検討を行った。また開発されたデジタルツインにおいては、設計業務効率等の改善とともに自動的かつ継続的な省エネ化や省CO2化、在室者の快適性向上といった効果が多様な建物で期待できる。