Three-dimensional morphological modelling of concrete using multiscale Poisson polyhedra.

Area of Science:

• Materials Science
• Civil Engineering
• Computational Modeling

Background:

• Understanding concrete microstructure is crucial for predicting material properties.
• Existing models may not fully capture the complex, random nature of aggregate distribution.
• Accurate representation of cement paste and aggregates is essential for material performance.

Purpose of the Study:

• To develop a novel random morphological model for concrete microstructures.
• To validate the model using experimental data and morphological measurements.
• To provide a tool for simulating and analyzing concrete's internal structure.

Main Methods:

• Acquisition of a 3D concrete image using microtomography.
• Development of a two-phase model (cement paste and aggregates).
• Modeling aggregates as a combination of scaled Poisson polyhedra.
• Generation of polyhedra packings using a novel algorithm.
• Validation through morphological measurements.

Main Results:

• Successful development of a random morphological model for concrete.
• The model accurately represents the two main phases of concrete.
• The algorithm effectively generates realistic polyhedra packings.
• Model validation confirmed accuracy through morphological measurements.

Conclusions:

• The developed random morphological model provides a robust representation of concrete microstructures.
• This model can aid in the design and analysis of concrete materials.
• The methodology offers a pathway for simulating complex material morphologies.