2022-06-18

In order to continue to do a good job in the recruitment of master/doctor graduate students of TSINGHUA SIGS Institution for Ocean Engineering (IOE) , and promote college students to understand the relevant situation of the institute and the recruitment of master/doctor majors, the 2023 Enrollment seminar of IOE was successfully held in the form of online live broadcast on June 11, 2022. Professor Chen Daoyi and teachers from all disciplines attended the meeting, and I presided over the meeting.

At the meeting, Professor Chen Daoyi introduced the general situation of IOE. Associate Professor Li Yi, Professor Wang Yong, Professor Zheng Xiangyuan and MYSELF, assistant Professor Shen Xinwei introduced the major directions, teaching staff and training programs of electronic information, resources and environment, civil and hydraulic engineering and energy dynamics. Mr. Zhou Jin, Mr. Wang Enhao and other faculty members answered questions from students about their professional background, source institutions, enrollment quota and employment destination. In addition, Zhang Xiangyu, 2020 outstanding student representative of civil and hydraulic engineering, shared with you his study and life in TSINGHUA Institute of Deep Research and extended his best wishes.

More than 67,000 people participated in the event through China Education Online and B live broadcasting platform, which further expanded the influence of IOE and will also facilitate the smooth implementation of 2023 postgraduate enrollment.

Note: The link to the presentation will be reviewed throughout: https://live.eol.cn/watch/3132915.

2022-06-16

Development of a micro-in-meso-scale framework for simulating pollutant dispersion and wind environment in building Groups has been published online in the Journal of Cleaner Production.

Simulation of pollutant dispersion and wind environment is of significant importance to the safety and health of pedestrians and residents. However, the high threshold of data collection and model configuration has hindered its application in the building industry. To address the problem, this paper proposes a Micro-in-Meso-scale (MiM) simulation framework for automatic and foolproof pollutant dispersion and wind environment simulation. The meso-scale Weather Research and Forecast (WRF) model is first carried out to obtain city-level meteorology data with sufficient resolution. The micro-scale Computational Fluid Dynamics (CFD) model is then initialized to simulate pollutant dispersion and wind environment in building groups. The automatic simulation platform of pollutant dispersion and wind environment was further developed. The platform encapsulates the professional knowledge required for wind environment simulation and can realize automatic configuration of model parameters according to simulation scenarios. Platform users do not need to pay attention to the complicated numerical model configuration and operation details, just specify the simulation time and place, can get three-dimensional visualization results of building wind environment under given conditions, reduce the professional threshold involved in wind environment simulation.The research results were supported by the National Natural Science Foundation of China, Shenzhen Key Laboratory of Marine Intelligent Sensing and Computing, and Shenzhen city planning and land resources committee.    

Note: Journal of Cleaner Production is an important SCI indexed journal in the field of environmental science and ecology, with an impact factor of 9.297 in 2021. The first author of this paper is Leng Shuo, a 2018 doctoral student in the Department of Civil Engineering of Tsinghua University. I am the corresponding author. Authors also include Assistant Professor Sun Wei Li, master student Wu Haoyang and Associate Professor Li Binbin at SIGS.

2022-06-15

Participated in the compilation of "City Information Modeling (CIM) Technology Research and Application" (April 2022 first edition)was officially published.

Editor of the book by the Guangzhou Municipal Housing and Urban-Rural Development Bureau, Guangzhou construction Science and Technology Center, Tsinghua university and other units participated in, China building industry publishing house, and from CIM development present situation, related technology, platform construction, application and development trend from four aspects to conduct a comprehensive introduction to the CIM technology, can be used as the government and enterprise resources for the development of CIM technology application.

2022-06-13

Congratulations to Min Yantao and An Peng on successfully defending their bachelor's thesis in the Department of Civil Engineering of Tsinghua University!

Among them, Min Yantao's thesis title is "Knowledge Driven Building Wind Environmental Disaster Assessment Technology". Accordingly, this paper builds a building wind environmental disaster assessment platform by collecting previous data, with functions as damage prediction, similar case recommendation, automatic question answering achieved, utilizing machine learning, nearest neighbor search, knowledge graph along with other technical methods. It serves for providing rapid prediction and assessment of the possible damage of buildings in a certain wind environment, and offering relevant knowledge as referenceto response strategies.

Anpeng's thesis title is "Data-Driven Knowledge Extraction in the Field of Highway Engineering Safety". The thesis is based on  the field of highway engineering safety management and driven by data, this research aims at applying knowledge extraction method to build a large-scale knowledge map and a professional term database within the domain, and providing intellectual knowledge support solutions based on knowledge graph.

2022-06-07

"Entity labeling method, module and device based on dictionary matching" has obtained the invention patent certificate (patent number: ZL202011079331.0).

The present invention discloses an entity labeling method based on dictionary matching, modules and apparatus, comprising: the entity words are arranged into an ordered dictionary from small to large; Establish a forward index bar F for entity words, and the ith element in F is the largest prefix of the i character before the entity word in the ordered dictionary; Insert the sentence s into the ordered dictionary by size, use the largest entity word smaller than the sentence s in the ordered dictionary as the largest common prefix base word w, calculate the same characters as the first x of s and w, and then obtain the maximum prefix of sentence s in the ordered dictionary; Use the maximum prefix annotation information to add the annotation information to the corresponding entity words of the sentence s and cut it out of the s, otherwise the first word in the s will be cut out, and the remaining part after the cut out will be the sentence s, repeat the above steps until s is an empty string. The present invention after a binary search can obtain a split entity, and then obtain pre-labeled results, by this pre-labeling method to reduce the user's repetitive labor.

Note: The inventors of this patent also include  Liu Yi, a doctoral student, and Wu Langtao, a master student.