2022-02-09

"Method and Device for Simulation and Prediction of Water Pollution Diffusion based on Concentration Analysis" has obtained Certificate of Invention Patent. (Patent No. ZL202111240610.5).

This invention provides a method and device for simulation and prediction of water pollution diffusion based on concentration analysis. It can realize efficient diffusion simulation of various pollutants in water, especially in a large range of water (such as the ocean). In this method, the simulated area is divided into grids according to small squares, and a numerical value is used to record the pollutant concentration at the center of each grid. The corresponding values of all grids constitute a concentration matrix representing pollutant distribution, which is used to calculate the variation of pollutant concentration in each grid. Among them, the change process of pollutant concentration is divided into three independent sub-processes: migration process, dispersion process and attenuation process. According to analyzing and calculating the change of concentration matrix caused by each sub-process, the sub-process is superimposed to obtain the diffusion process of pollutants, the concentration matrix is updated. According to the time step the concentration matrix is iteratively calculated, and the pollutant concentration in the corresponding grid region is represented by the numerical value in the concentration matrix. Thus, it can realize the simulation and prediction of pollutant diffusion process.

Note : This patent is the inventor of Liu Yi, Guo Xueqing, Zhang Jianmin, Li sunwei, and myself.

2022-02-08

"Method and Device for Simulation and Prediction of Water Pollution Diffusion based on Particle Motion" has obtained Certificate of Invention Patent. (Patent No. ZL202111232257.6).

This invention provides a method and device for simulation and prediction of water pollution diffusion based on particle motion. It can realize efficient diffusion simulation of various pollutants in water, especially in a large range of water (such as oceans). Thus, it can solve the problems of analysis and prediction of pollutant diffusion process, concentration distribution and diffusion path. In this method, the pollutants in the simulated area are decomposed into a large number of pollutant particles, and the motion state change of each pollutant particle can be calculated. Among them, the motion state change process of pollutant particles is decomposed into three independent sub-processes: directional motion, random motion and random attenuation. Through the separate calculation and analysis of each sub-process, the whole diffusion process of pollutant particles is obtained by superimposing the sub-processes. The diffusion process of pollutant particles is iterated continuously, the pollutant particles in the sub-region of the simulation region are counted at the selected time point, and the pollutant concentration at the corresponding place is represented by the number of particles in each sub-region.

Note : This patent is the inventor of Liu Yi, Guo Xueqing, Zhang Jianmin, Li sunwei, and myself.

2022-01-12

The annual summary meeting of the "13th Five-Year Plan" National Key R&D Program Project of China" Rural Housing Design and Construction Technology Application Information Platform" was successfully held online.

The meeting summarized the overall promotion of the project in 2021 and made arrangements for the work plan for 2022. The research group reported on the progress of the research task "Lightweight Platform Framework and Interface Technology Based on GIS/BIM". At present, the research mission is proceeding smoothly as planned. In 2021, the research task conquered the two key technologies of bim model lightweight and efficient transmission, as well as web-based GIS/BIM model visualization. In 2022, the focus of the research task includes the development and testing of the platform framework, and the completion of the work related to the conclusion of the project.

2022-01-04

Rule-based information extraction for mechanical-electrical-plumbing-specific semantic web has been published in the journal Automation in Construction.

This paper proposes a rule-based approach for professional mechanical, electrical and plumbing (MEP) Information extraction (IE) task, including a “snowball” strategy to collect large-scale MEP corpora, a suffix-based matching algorithm on text segments for named entity recognition (NER), and a dependency-path-based matching algorithm on dependency tree for relationship extraction (RE). Based on the above method, the corpus containing 65MB of text is constructed, and the algorithm is used to automatically construct a corpus including 15978 entities, Knowledge graph of 65110 relationships. The test shows that the proposed method has an accuracy of 81% in identifying named entities and 75% in relation extraction, which is 37% and 49% higher than that of existing deep learning methods.

Note: Automation in Construction, short as AUTCON, is the most influencial academic journal in the field of information technologies in civil and building engineering.It belongs to the Q1 district SCI journal, with a 2021 impact factor of 7.7. The research results have been funded by the National Natural Science Foundation of China and the Glodon BIM Joint Research Center.

2021-12-26

At the 7th National BIM Academic Conference, the research group published three papers. All three papers were included in the Proceedings of the 7th National BIM Conference.

Among them, Zhang Yunyi, a postdoctoral fellow in the Department of Civil Engineering of Tsinghua University, published "Big Data Management Mode and Architecture of Energy Consumption During Building Operation and Maintenance Period" for the first author, which is mainly based on the unified management energy consumption data management mode, and proposes a data management and application architecture for energy consumption during building operation and maintenance. Wu Langtao, a 2019 master's student at the Institute of Disaster Prevention and Mitigation of the Department of Civil Engineering of Tsinghua University, published "Construction and Application of Knowledge Graph of Building Mechanical and Electrical Equipment" for the first author, which proposes a method of using knowledge graph to construct a knowledge retrieval application framework. Lin Chao, a 2020 master's student at the Institute for Ocean Engineering, Tsinghua Shenzhen International Graduate School, published "Research and Classification of Intelligent Supervision Informatization Enabling Technology" as the first author, which comprehensively investigated and sorted out the intelligent supervision informatization technology.