Comparison of three algorithms generating virtual microstructures in terms of the degree of randomness

• Published in: Composites. Part A, Applied science and manufacturing Vol. 177; p. 107959
• Main Authors: Ding, Yongfeng, Camanho, P.P., Silva, Arlindo
• Format: Journal Article
• Language: English
• Published: 01.02.2024
• Subjects: algorithms; quantitative analysis
• ISSN: 1359-835X
• DOI: 10.1016/j.compositesa.2023.107959

The degree of randomness of fibre arrangements within the resin area is of great importance to the composites’ micro-distribution of stress and strain. A new algorithm, hard-core model & random shaking model, is developed to generate the unidirectional continuous fibre-reinforced composites. Another two algorithms, the random sequential expansion model and the initially periodic shaking model are presented and improved to generate the representative volume element microstructures. Statistical analyses are performed to compare these representative volume element microstructures generated by different algorithms at different fibre volume fractions. A quantitative approach is first applied to provide the exact degree of randomness of both virtual and real microstructures. Results of qualitative and quantitative analyses show that the novel algorithm is capable of generating statistically equivalent fibre distributions to real continuous fibre-reinforced composites.