Zhenzhong Hu

    He received both his BE and PhD degree in the Department of Civil Engineering at Tsinghua University, China. He was a visiting researcher in Carnegie Mellon University.
    He is now the associate professor in Shenzhen International Graduate School, Tsinghua University, and also the secretary general of the BIM Specialty Committee of the China Graphics Society.
    His research interests include information technologies in civil and marine engineering, building information modeling (BIM) and digital disaster prevention and mitigation.
  • 2023-01-17

    Research on semi-automatic geo-referencing approach in city information modeling has been published in the Journal of Graphics.

    The 3D model lacking spatial reference needs geo-referencing in the Geographic Information System (GIS) when City Information Modeling (CIM). Based on a geometric reference point, semi-automatic geo-referencing can be achieved while ensuing accuracy. After the 3D model is imported into GIS, the local coordinate system (LCS) is transformed to the surface of the geodetic ellipsoid, based on any reference point near the model. Afterwards, the geographic reference point on the 3D model is picked up in GIS, and the LCS coordinates are calculated according to the current transformation matrix. Combined with the geodetic coordinates of the reference point, the correct transformation matrix of the model can be obtained. When there is no geographic reference point, artificial geo-referencing can be achieved by continuously estimating the geographic reference point based on this approach. The semi-automatic georeferencing approach is realized by Cesium, and is applied using a 3D model from an actual project, which proves the feasibility of the approach. The above approaches rely on the spatial transformation algorithms from LCS to ECEF coordinate system using a single reference point. The algorithms can be divided into basic approach, reference point projection approach, origin projection approach and approximate origin approach according to the features and applicable scenarios. The latter 3 algorithms are applied in verification part, and based on the calculated results, the influence of the distance between the reference point and the LCS origin on these algorithms is analyzed.

  • 2023-01-05

    On the afternoon of January 4th, Zou Dong, Secretary of the Party Committee of Guangzhou Metro Construction Management Co., Ltd. and a delegation visited Tsinghua Shenzhen International Graduate School to exchange scientific research work with members of the research group.

    At the meeting, Chen Xiangxiang, a member of the research team, reported the research application and prospect of coding, open platform, IFC and other technologies in the field of rail transit. This paper makes a supplementary introduction to abstract terms such as component, construction resource, group and resource layer involved in the conceptual model of construction management, and further introduces the research progress of the research group on digital twinning and knowledge graph. Later, Secretary Zou Dong had an in-depth technical exchange with the members of the research group, and affirmed the ability of our research group to conquer the core key technologies of the intelligent construction of rail transit. He also expected that both sides could continue to strengthen cooperation, innovate and make greater achievements on the basis of the past!

    Since 2014, the research group and Guangzhou Metro Construction Management Co., LTD., after eight years of joint research, established the information theory and technology system of digital twins-driven digital construction of rail transit engineering and the design standards and methods for digital construction of rail transit. Put forward and practice the digital twin driven rail transit digital integrated construction solution and management mode; A BIM convergence platform with independent property rights for rail transit engineering construction has been developed.

  • 2022-12-01

    The software copyright of "Model Visualization Platform Based on Web" has been obtained.

    The platform is one of the landmark achievements of the national key research and development Plan of the 13th Five-Year Plan, the Information Platform for the Application of Rural Housing Design and Construction Technology. This topic aims at the problem of the backward application level of information technology in the process of rural housing design and construction. Combined with the practical engineering needs of rural housing, it provides a series of theories, methods and technical support for the information transformation of rural housing. As the carrier of integration, integration and application of various technologies, "Web-based model visualization platform" has become a fundamental role in the subject. The basic functions of the platform include the transformation and integration of multi-source heterogeneous rural housing data, lightweight and efficient transmission of rural housing model, and high-efficiency multi-scale visualization based on the Web end, so as to realize the integration and unified external service of rural housing design and construction information, and provide an ideal platform for the communication and collaboration of villagers, designers and construction personnel. On this basis, the platform provides standardized data interface and development interface, access to a series of rural housing information applications such as site selection planning evaluation, atlas inquiry recommendation, construction and construction simulation, etc., so as to provide all-round information technology support for the design and construction of rural housing.

  • 2022-11-29

    Localization and topological observability analysis of a moored floating structure using mooring line tension measurements have been published online at Ocean Engineering.

    Mooring is a widely used stationkeeping approach for floating structures. The main contribution of this paper is to propose a novel localization scheme, which provides additional real-time position references and estimates the unknown anchor positions to a moored structure simultaneously using tension measurements from the mooring monitoring system. The proposed method enhances the safety of moored floating structures by extracting redundant position information from tension measurements instead of using satellite signals. A line-of-sight model that transforms tension measurements into range signals is developed. Fairleads, turret dynamics, and slow-varying current profiles are considered to provide a more realistic and robust tension-based localization scheme than early research. A distance-only sensor network is constructed, and graph rigidity theory is applied to analyze its topological observability. Two simplified models are presented for some specific applications: (1) to find the broken anchors offline with stored data for any vessels equipped with tension cells and (2) to position the moored structures with known anchor positions. Sensitivity studies show that the influence of unknown surface current is small, while the effects of unknown current profiles are larger. The positioning accuracy can be improved by increasing the number of virtual vessels.

    Note: Ocean Engineerin belongs to the top of the field of engineering and technology magazine, with a 2022 impact factor of 4.372.

  • 2022-11-11

    A hybrid data mining approach on BIM-based building operation and maintenance has been cited more than 100 times.

    Huge amounts of data are generated daily during the operation and maintenance(O&M)phase of buildings. These accumulated data have the potential to provide deep information that can help improve facility management . Building information Model/Modeling(BIM) technology has proven potential in O&M management in some studies, making it possible to store massive data. However, the complex and non-intuitive data records, as well as inaccurate manual inputs, raise difficulties in making full use of information in current O&M. Aiming at the problem that the data format of large buildings in the operation and maintenance period is complex and cannot be fully utilized, this study proposes a set of hybrid data mining methods which can accurately obtain useful information and detect abnormal operation and maintenance records. This paper is one of the first papers to report the application of data mining algorithm on BIM data set and service for large-scale construction operation and maintenance projects. The mixed mining method for operation and maintenance data has strong universality, so it has been widely concerned and cited after publication, including a number of review papers in the top issue of SCI.

    Note: Building and Environment, the top journal in the field of Engineering and Environmental science, has an impact factor of 7.093 at present. The first author of this paper is Peng Yang, a 2015 Master's student in the Department of Civil Engineering of Tsinghua University. I am the corresponding author.