Advanced microstructural analysis of cement-based materials: Integrating x-ray computed tomography and deep learning for enhanced crack growth understanding

Research output: Contribution to conferencePaperResearchpeer-review

Abstract

Understanding the mechanism of crack growth in cement-based materials under mechanical loading involves complex interactions between microstructural components, including aggregates, voids, and cement paste. This paper presents a unique approach that combines X-ray computed tomography (XCT) with deep learning to segment these components precisely. By leveraging XCT's high-resolution 3D imaging capabilities and the robustness of deep learning algorithms, our method provides a detailed characterization of the microstructure of cement-based materials. This detailed structural information is crucial for understanding crack initiation and propagation processes, ultimately contributing to developing more durable and sustainable concrete. Our results highlight the significant potential of deep learning in enhancing our understanding of damage and failure mechanisms in cement-based materials, providing valuable insights that can lead to improved material performance and longevity.
Original languageEnglish
Publication date2025
Number of pages7
DOIs
Publication statusPublished - 2025
Event12th International Conference on Fracture Mechanics of Concrete and Concrete Structures - TU Wien, Vienna, Austria
Duration: 23 Apr 202525 May 2025
https://framcos12.conf.tuwien.ac.at/

Conference

Conference12th International Conference on Fracture Mechanics of Concrete and Concrete Structures
LocationTU Wien
Country/TerritoryAustria
CityVienna
Period23/04/202525/05/2025
Internet address

Keywords

  • Fatigue damage
  • Cement-based materials
  • Deep learning

Fingerprint

Dive into the research topics of 'Advanced microstructural analysis of cement-based materials: Integrating x-ray computed tomography and deep learning for enhanced crack growth understanding'. Together they form a unique fingerprint.

Cite this