2023-09-22

"A passive earth pressure model test device for finite viscous soil behind a flexible retaining wall of foundation pit" has obtained the invention patent certificate (patent number: ZL201911214637.X).

The invention provides an active and passive earth pressure model test device for finite viscous soil behind a flexible retaining wall of a foundation pit, which comprises: a test bench unit comprising a fixed steel frame, a model box embedded in a fixed steel frame and a transparent movable baffle plate. The model box is equipped with a model viscous soil with a ratio similar to the weighted average value of physical and mechanical properties of the prototype viscous soil; The loading and unloading unit comprises at least five displacement control handles, a reaction device and an associated model wall. The displacement control handle is uniformly arranged in the buried depth direction of the model wall. One side support of the displacement control handle is fixed on the reaction device, and the other side is fixed with the model wall. Contact and non-contact monitoring unit, contact monitoring unit includes: micro earth pressure gauge, micro displacement meter; The non-contact monitoring unit includes a PIV monitoring unit. The law and magnitude of the passive earth pressure and displacement of soft viscous soil behind the wall of flexible enclosure structure can be obtained under various limited soil widths.

2023-09-09

On September 9, I was invited to Laoshan Laboratory to attend a meeting on Marine environmental monitoring, and made a report on the work and achievements of the research group in related aspects. In addition, Li Sunwei, a core member of the research team, and Liu Yi, a doctoral student, also participated in the meeting.

Laoshan Laboratory is a breakthrough, leading and platform integrated new scientific research institution in the Marine field approved by the State. It is jointly built by the State ministries, Shandong Province and Qingdao, aiming to focus on the innovation-driven development strategy and the overall requirements of building a maritime power, adhere to the "four aspects", carry out scientific and technological innovation and institutional innovation, and strive to build a national Marine strategic scientific and technological force.

2023-08-30

Approaches Improving Cesium Rendering Performance for Displaying Massive 3D Models has been published in the Journal of Information Technology in Civil Engineering and Architecture.

Rendering performance optimization is needed to display dense 3D models in Cesium requires improving the. However, the native approaches are unstable and will cause bad rendering results. To avoid this problem as much as possible, a mechanism is proposed in this paper to enable high-performance rendering when the camera state changes in which the camera states change pauses primitives loading, and applies parameter settings and algorithms that can improve rendering performance but may cause bad rendering results. Besides, this paper proposes a heap-based 3D tile draws command limitation approach, which can directly limit the number of draw commands and reduce the amount of GPU computation. The results prove that the mechanism of high-performance rendering when camera state changes are feasible. The mechanism outperforms the max screen space error (MSSE) increasing and the draw commands limiting, which excels reducing the resolution scale (RS) in improving rendering performance. Limiting the draw commands can ensure the stability of rendering performance compared with increasing MSSE.

Note: The first author of this paper is Wang Hengwei, a postdoctoral fellow, and I am the corresponding author. The research results have been funded by the National Key Research and Development Program of China and the National Natural Science Foundation of China.

2023-08-28

The 19th China CAE Engineering Analysis Technology Conference and the 5th China Digital Simulation Forum were successfully held in Xiamen, Fujian Province, from August 18th to 20th. The theme of this year's conference was "Simulation Leading Intelligent Manufacturing, Digital Driving Twinning."

Among them, the 18th was the keynote report of the conference, I attended the conference, and made a special invited report on the topic of "The fundamental way to achieve China's 'double carbon' goal", introducing the current situation of China's "carbon neutrality". On the evening of the 18th, the award ceremony of the "Yundao Cup" China Digital Simulation Competition was held. The team of the research group won the second prize of the university group of the Digital simulation Innovation Competition, and the paper submitted by the research group "The Application research of digital twin in the installation of offshore wind power" won the excellent paper award. On the 20th, members of the research group participated in the offshore engineering and petroleum equipment simulation application Forum, and reported on the submitted papers, demonstrating the application feasibility of digital twins in offshore wind power installation.

The China CAE Engineering Analysis Technology Annual Conference is currently the largest and most influential professional exchange event in our country, with the highest level of technology. The conference was attended by more than 800 offline and more than 8,000 online representatives from scientific research institutes, enterprises, universities and major media across the country, providing a valuable platform for strengthening research, sharing and cooperation in the field of CAE.

2023-08-25

"Prediction of Magnitude and Epicentral Distance Based on Neural Network" has been published in the journal of Municipal Technology.

In order to predict rapidly the magnitude and epicentral distance, various machine learning techniques including convolutional neural network, fully connected neural network, GELU activation function, dropout and multi-task learning were applied to establish an efficient model for predicting the magnitude and epicentral distance. The model was trained, validated, and tested by the Stanford Earthquake Dataset （STEAD）. The predicted values demonstrated a better positive correlation with the actual ones, with determination coefficients of 0.938 and 0.881 for magnitude and epicentral distance respectively. These results provides new ideas and methods for the development of earthquake warning systems.