World Congress on

Civil, Structural, and Environmental Engineering

  • Renaissance London Heathrow Hotel
  • March 10-11, 2025
;
Dr. Mingzhi Wang

 

Dr. Mingzhi Wang

Harbin Institute of Technology
China

Abstract Title: Image-based Simulation of the Mesoscopic Environment-material Interaction

Biography:

Dr. Wang studied Civil Engineering at the Harbin Institute of Technology (HIT) and the University of Birmingham. He received his double B.Eng in 2012. He then joined the Geotechnical and Environmental Research Group under the supervision of Prof. Abir Al-Tabbaa at the Engineering Department, University of Cambridge. After receiving his Cantab PhD degree in 2017, he began to conduct research and teaching under the supervision of Prof. Wei Wang in the School of Civil Engineering, HIT. He is now serving as an Associate Professor at HIT. He has published more than 20 research articles in SCI (E) journals.

Research Interest:

The latest development of 2D and 3D image recognition suggests a new path to investigate environment-material interaction during remediation. Porous media modelling is now available through image-based reconstruction, which provides a realistic boundary condition for chemical processes including dissolution, precipitation and reaction. This presentation demonstrates the algorithmic techniques including image segmentation, particle packing, computational fluid dynamics and computational chemistry in mesoscopic environment-material interaction.  An innovative architecture is presented to efficiently establish a natural rock database with individual 3D printable volumes and surfaces. The database can be used to perform heterogeneous modelling with realistic aggregate input and distributive analysis of a specific constituent of interest. The necessity of customizing computer vision in the application of concrete composite is discussed with evidenced digital damage in the non-destructive measurement. A comparison with available reconstruction methods with feasibility is performed to demonstrate that characterizing practical porous media with detailed information on each constituent can provide a more realistic representation of the composite.  The feasibility of such a proposal is evidenced by a discrete particle packing simulation scheme, which enables the generation of random 3D pore structures. The discrete packing model adopts a pre-existing algorithm and an improvement of pseudo-contact mechanics is introduced. An image-based model of heavy metal immobilization in wooden media is further presented as a well-established example. Digital testing methods are presented to calculate the surface area, specific surface area, density, porosity and immobilization amount.  Three types of media are constructed for a comparison of immobilization capacity and efficiency. The computational approach provides a quantitative evaluation of the immobilized ions in arbitrary media. The proposed digital testing is feasible for both experimentally obtained images and structures from algorithm generation. In the end, the perspective of artificial intelligence is generally discussed based on data compatibility.

Keywords: Porous Media; Heavy Metal Immobilization; Heterogeneous Model; Image Segmentation; Multiphase Fluid.

Upcoming Conferences

World Summit on Renewable & Non-Renewable Energy

Read More...

3rd World Congress on Materials Science & Engineering

Read More...