TY - JOUR
T1 - Modelling method of fibre distribution in steel fibre reinforced concrete based on X-ray image recognition
AU - Li, Yue
AU - Ruan, Xin
AU - Akiyama, Mitsuyoshi
AU - Zhang, Mingyang
AU - Xin, Jiyu
AU - Lim, Sopokhem
N1 - Funding Information:
The support from the National Key R&D Program of China (Grant number 2018YFB1600100 ), the National Natural Science Foundations of China (Grant numbers 51678435 , 52078367 ) and the Peak Discipline Construction Projects of Civil Engineering of Tongji University is gratefully acknowledged. The opinions and conclusions presented in this article are those of the authors and do not necessarily reflect the views of the sponsoring organizations.
Publisher Copyright:
© 2021 Elsevier Ltd
PY - 2021/10/15
Y1 - 2021/10/15
N2 - The accurate and efficient simulation of steel fibre reinforced concrete (SFRC) is of great significance for its further application in civil engineering. Problems of existing modelling methods can be recognized as two points; (1) the integral model only simplified fibre contribution into the unified post-peak ductility in concrete element but cannot reflect the material variation caused by real fibre distribution, and (2) the separate model built the concrete matrix and fibre inclusions respectively, but the directly random fibre distributions still need to be refined based on the actual material. In this paper, based on the X-ray image recognition results from real SFRC material, a mathematical model of fibre distribution and orientation is proposed for SFRC simulation, and the detailed value distributions of geometrical parameters are optimized for convenience and accuracy of material modelling. In the geometrical extraction of near 4000 fibres from X-ray images, the whole comprehensive procedure of image processing, enhancement and recognition is presented with technologies like pixel semantic classification and Hough transform. With the proposed geometrical mathematical model, the modelling method that consists of multiple refined sampling is illustrated with mechanical applications. After the model verification by experimental comparison, the detailed contribution of steel fibre and the effect of fibre distribution refinement are discussed as well.
AB - The accurate and efficient simulation of steel fibre reinforced concrete (SFRC) is of great significance for its further application in civil engineering. Problems of existing modelling methods can be recognized as two points; (1) the integral model only simplified fibre contribution into the unified post-peak ductility in concrete element but cannot reflect the material variation caused by real fibre distribution, and (2) the separate model built the concrete matrix and fibre inclusions respectively, but the directly random fibre distributions still need to be refined based on the actual material. In this paper, based on the X-ray image recognition results from real SFRC material, a mathematical model of fibre distribution and orientation is proposed for SFRC simulation, and the detailed value distributions of geometrical parameters are optimized for convenience and accuracy of material modelling. In the geometrical extraction of near 4000 fibres from X-ray images, the whole comprehensive procedure of image processing, enhancement and recognition is presented with technologies like pixel semantic classification and Hough transform. With the proposed geometrical mathematical model, the modelling method that consists of multiple refined sampling is illustrated with mechanical applications. After the model verification by experimental comparison, the detailed contribution of steel fibre and the effect of fibre distribution refinement are discussed as well.
KW - Image recognition
KW - Modelling method
KW - Steel fibre reinforced concrete
KW - X-ray
UR - http://www.scopus.com/inward/record.url?scp=85109623737&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85109623737&partnerID=8YFLogxK
U2 - 10.1016/j.compositesb.2021.109124
DO - 10.1016/j.compositesb.2021.109124
M3 - Article
AN - SCOPUS:85109623737
SN - 1359-8368
VL - 223
JO - Composites Part B: Engineering
JF - Composites Part B: Engineering
M1 - 109124
ER -