| Abstrakt: |
This study investigates the residual compressive strength and modulus of elasticity of recycled aggregate concrete containing ground granulated blast furnace slag (RAC-GGBS) exposed to high temperatures, presents the numerical optimization of the mix design parameters of RAC-GGBS to achieve maximized residual compressive strength, and develops prediction models for residual compressive strength of RAC-GGBS based Gene expression programming (GEP) approach. The considered variables were the percentage replacement of RCA (50, 75, 100%), percentage replacement of GGBS (0, 20, 40, and 60%), water to binder ratio W/B (0.4 and 0.5), compressive strength (fcu) and exposed temperatures of 25 ℃ (room temperature), 200, 400, 600, and 800 ℃. The experimental results show that addition of GGBS significantly enhanced the residual compressive strength and modulus of elasticity of RAC-GGBS while the SEM analysis conducted revealed a more compact microstructure with fewer pores and micro-cracks and better stabilized C–S–H hydrate formed at ITZ of RAC containing GGBS compared to RAC without GGBS after exposure to high temperature. The optimum condition obtained for RAC-GGBS exposed to high temperatures was 99.8%, 46.0% and 0.45 for RCA, GGBS and W/B, respectively, having maximized residual compressive strengths of 24.0, 25.5, 22.4, 17.2, 7.3 MPa after 25 ℃, 200 ℃, 400 ℃, 600 ℃ and 800 ℃ temperatures, respectively. The predictive ability of GEP model is high with R2values above 0.97 and low measured statistical errors. The mean absolute error (MAE) values of 0.889 and 1.244 and root mean square error (RMSE) of 1.064 and 1.502 for the training and validation datasets respectively, validate the GEP model's high accuracy and strong capacity to predict the residual compressive strength of RAC-GGBS. |
Full text from SpringerLink