Simulation and parameter prediction model of rheological properties of fiber reinforced concrete

Abstract

For higher concrete strength, fiber-reinforced concrete is a hot research topic. In the transportation process of fiber-reinforced concrete, flow characteristics are critical indicators of pumping fiber-reinforced concrete. In this paper, firstly, the rheological characteristics of steel fiber-reinforced concrete are measured based on the small slump cylinder. The rheological parameters, such as slump and expansion of steel fiber reinforced concrete with different proportions, are analyzed, and the rheological properties of the slurry are tested by rheometer. Secondly, to realize the effective prediction of the rheological parameters of steel fiber reinforced concrete, the Fluent numerical simulation software in computational fluid dynamics (CFD) is used to simulate the rheological properties of concrete in the micro slump cylinder. By comparing with the experimental results, the simulation results are in good agreement with the actual results, and the maximum error is only 2 mm, which verifies the accuracy of the numerical model. Finally, based on the test results of 25 groups of rheological parameters of fiber-reinforced concrete as the original data set, the prediction model of rheological parameters of fiber-reinforced concrete based on Sparrow Search Algorithm - Xtreme Gradient Boosting (SSA-XGBoost) is established. The prediction results show that SSA-XGBoost has high applicability in the prediction of the dynamic yield stress of the mixture, and the experimental value R2 is as high as 0.99.

Keywords

• Concrete
• Steel Fiber
• Simulation
• Rheology
• Machine Learning