2025-05-24

The paper "Fluid flow and heat transfer characteristics of manifold microchannel heat sinks with ribs of different shapes" has been published in the Journal International Journal of Heat and Mass Transfer.

Manifold Microchannel Heat Sink (MCHS) is an effective device to dissipate the heat generated from the chips with large-scale circuits. Although it is found that adding ribs to the microchannels of the traditional MCHS improves its performance in heat dissipation, there is no systematic study to investigate the effect of adding ribs to the manifold MCHS. The pressure drops, mean temperatures, the Nusselt number, the heat enhancement efficiency, the characteristics of flows and heat transfers have been studied numerically in the manifold MCHS with five different shaped ribs and compared to the smooth manifold MCHS in this paper. The comparison shows that the development of boundary layers in the flow and temperature fields is interrupted by the ribs, which leads to an enhancement in heat dissipation. Compared to the smooth manifold MCHS, the ribs reduce the temperature of the substrate bottom by 9.2%-25.1% and the Nusselt numbers are increased by 11.5%-45.2%. However, the ribs in the microchannels increase the pressure drops by 55.7%-111.4%. As the Reynolds number increases, the influence of ribs on these key indicators becomes more pronounced. Through the computational analysis of the thermal enhancement efficiency, it is found that benefits of adding ribs overwhelms at higher Reynolds number (exceeds 200). The elliptical ribs are recommended among the different rib shapes under investigation and the heat enhancement efficiency achieves a maximum value of 1.17 when the Reynolds number is 450.

Note: The International Journal of Heat and Mass Transfer belongs to the Q1 zone of engineering and technology journals, with an impact factor of 5.0. The first author of the paper is Chen Shoujun from the Institute for Ocean Engineering. Teacher Li Sunwei is the corresponding author. The research results were supported by the Guangdong Basic and Applied Basic Research Foundation.  The authors acknowledge the support from the Tsinghua Shenzhen International Graduate School - Shenzhen Pengrui Young Faculty Program of Shenzhen Pengrui Foundation.

2025-05-16

Congratulations to Liu Longxiang on successfully passing the master's thesis defense at Tsinghua University!

The title of Liu Longxiang's paper is "Research and Application of Tunnel Multi-Source Heterogeneous O&M Data Fusion Based on Ontology". This research takes multi-source heterogeneous operation and maintenance data of tunnels as the research object, introduces ontology semantic web technology, and integrates and utilizes various tunnel operation and maintenance data resources. Firstly, this study constructed a classification system for tunnel operation and maintenance data. For different types of tunnel operation and maintenance data, domain ontology models were established respectively in different ways to achieve the unification of data formats. Subsequently, this study proposed an ontology mapping strategy based on comprehensive similarity to achieve efficient and accurate mapping among ontology models of different operation and maintenance data. Finally, a healthy operation and maintenance strategy for tunnel structures is proposed, including steps such as application ontology construction, mapping and semantic query, structural health assessment, and maintenance strategy selection. In addition, this study also developed a multi-source heterogeneous operation and maintenance data management platform for tunnels and applied it in the actual engineering project of Tanglangshan Tunnel in Shenzhen City, achieving good results.

2025-05-12

On May 12th, the 50th International Exhibition of Inventions  of Geneva  announced the list of winners. The invention "ChatTwin: A Digital Twin Platform for Virtual and Real Verification and Dynamic Monitoring of Construction Processes Empowered by Large Models" by the teaching team from the Department of Civil Engineering of Tsinghua University won the gold award.

The project team has invented a large language model-driven engineering digital twin construction platform and technology - ChatTwin. ChatTwin involves a series of key technologies such as "understanding - reasoning - generation - interaction", and is simple, easy to use, intelligent and efficient. By constructing large language models in vertical fields, precise understanding of professional texts/languages in the construction field, efficient reasoning and calculation of complex engineering logic, as well as efficient analysis and processing of massive engineering data have been achieved. Thus, the learning and usage costs of complex engineering systems can be significantly reduced, and the BIM design review time can be effectively shortened, as well as the risk of problem omission. This system also integrates engineering data from multiple different software or systems, enabling dynamic identification and tracking of potential quality and safety risks during the construction phase.

The ChatTwin completion team also includes teachers Lin Jiarui, Lu Xinzheng, Pan Peng, and Zhang Jianping from Tsinghua University, students Zheng Zhe, Zhou Yucheng, Song Shengyu, Chen Keyin, Ni Xiangrui, Han Jin, and Liao Wenjie, as well as Wu Dapeng, the general manager of Beijing Yunjianxin Technology Co., LTD.

Note: The International Exhibition of Inventions  of Geneva is one of the longest-running and largest invention exhibitions in the world. Jointly organized by the Swiss Federal Government, the government of the Canton of Geneva, the City of Geneva and the World Intellectual Property Organization, it is renowned for its strict selection and professional standards and is ranked first among the "World's Three Major Invention Exhibitions".

2025-04-17

The paper Vision-based adaptive cross-domain online product recommendation for 3D design models has been published in the journal Computer-Aided Civil and In Infrastructure Engineering.

Three-dimensional (3D) digital design is extensively adopted in the architecture, engineering, consulting, operations, and maintenance (AECOM) industry to enhance collaboration among stakeholders. Although recommendation systems are commonly employed to facilitate purchasing in e-commerce websites, none involves recommending online products to users from 3D building design models due to dimensional and stylistic discrepancies. This study proposes a visionbased adaptive cross-domain online product recommendation method, VacRed, for 3D building design models. First, a cross-domain approach is proposed to transform design models into e-commerce images, addressing discrepancies in dimension and style between them. Second, an adaptive mechanism is introduced to solve the issue of image quality instability caused by variations in generator weights during the training process of generative models. Third, a cross-domain product recommendation system is developed based on deep learning to recommend the top k relevant online products for a given building design product. Finally, experiments were conducted to ascertain the effectiveness of the VacRed method. The experimental results of this method demonstrate its excellent performance, achieving a precision rate (PR) of 87.20% and a mean average precision of 83.65%. This study effectively connects two main stages in the AECOM industry, design and purchasing, and two large communities, design and e-commerce.

Note: Computer-Aided Civil and Infrastructure Engineering belongs to the Q1 zone of engineering and technology journals, with an impact factor of 8.5. The first author of the paper is Zhou Xiaoping from Beijing University of Civil Engineering and Architecture. The research results were supported by the National Natural Science Foundation of China.

2025-04-12

Thesis AI-based prediction of seismic time-history responses of RC frame structures considering varied structural parameters has been published in the Journal of Building Engineering.

In this paper, an end-to-end framework for Intelligent Seismic Response Prediction, ISRPnet, is introduced. ISRPnet comprises a structural parameter module for discretizing reinforced concrete frame structures into a series of static features and an encoder-decoder architecture for encoding seismic loads and autoregressively predicting seismic responses. The model is trained on a data set of 16,544 cases generated through validated fibre-based finite element models. ISRPnet achieves promising performance on both frequent and rare earthquakes. ISRPnet rapidly and highly precisely predicts temporal responses for frequent earthquakes. The peak displacement predictions remain accurate for rare earthquakes. The superiority of the physical loss and the advantages of gated recurrent unit over long short-term memory are analysed in comparative experiments. Verification with unseen seismic waves beyond the training data shows the robust generalization and extrapolation capabilities of the framework. The proposed model accomplishes efficient surrogate computation of the full-process seismic response for a class of RC frame structures.

Note: Journal of Building Engineering, a journal in the field of engineering and technology, belongs to the SCI journal in Q1 region, with an impact factor of 6.7 in 2024. The first author is Ge Kang,  a master's student in 2022, and the corresponding author is Wang Chen, an assistant professor in the Department of Civil Engineering at Tsinghua University. The research results were supported by the National Natural Science Foundation of China and the Cross-disciplinary Research and Innovation Fund Research Plan of Tsinghua Shenzhen International Graduate School.