|
| Influence of fiber size effect on mechanical properties of deep shaft liner concrete |
|
Received:May 20, 2023
Revised:May 30, 2023
Accepted:June 27, 2023
Published Online:March 24, 2025
|
| View Full Text View/Add Comment Download reader |
| DOI： |
| KeyWord:steel fiber reinforced concrete; tensile strength; digital image correlation; mechanical properties, random forests |
| Author | Institution |
| Wang Xiaodong |
CINF Engineering Co., Ltd |
| WANG Yongbing |
Yunnan Chihong Zn & Ge Co., Ltd. |
| FANG Xugang |
Yunnan Chihong Zn & Ge Co., Ltd. |
| LI Diyuan |
School of resources and safety engineering, Central South University |
| JIANG Jingtai |
School of resources and safety engineering, Central South University |
| Wang Mimi |
School of resources and safety engineering, Central South University |
| Luo Pingkuang |
School of resources and safety engineering, Central South University |
|
| Hits: 526 |
| Download times: 209 |
| Abstract: |
| To explore the impact of fiber size effect on the mechanical properties and strain evolution of deep shaft liner SFRC, three concrete formulations with different steel fiber sizes and admixtures were designed. After testing the tensile and compressive mechanical properties of the samples, it was found that with the increase of steel fiber size and admixtures, the uniaxial compressive strength, tensile strength, flexural strength, and elastic modulus of the samples increased. Using non-contact digital image correlation technology to monitor the damage process of the sample, it was found that the addition of steel fibers significantly increased the number and length of surface cracks when the sample was damaged, compared to plain concrete samples. With the increase in the size of steel fibers, the width of surface cracks decreased when the sample was damaged, and the axial deformation of the concrete sample decreased. When the sample reaches the peak load, the plain concrete quickly loses stability and fails, while the steel fiber reinforced concrete sample still has a certain residual strength when it reaches the peak load. At the same time, the random forest model is used to train, test, and predict the testing strength of steel fiber reinforced concrete. The results show that the training, testing, and prediction sample accuracy is high, with R2>0.8, and the root mean square error value is small. This indicates that the improved random forest model is more accurate in predicting the strength of steel fiber reinforced concrete. Using this prediction model, the labor and material costs required for optimizing the steel fiber reinforced concrete formula can be greatly reduced, It can provide a reference for predicting the strength of concrete formulations in field engineering. |
| Close |
|
|
|