Abstract
AbstractPorous concrete plays a crucial role in addressing various environmental challenges and mitigating the impacts of climate change. It proves effective in reducing issues such as flooding, heat phenomena in the earth, and groundwater decline. Typically devoid of sand content, porous concrete’s key attributes lie in its permeability and compressive strength. Accurate prediction of these properties is essential for cost and time savings, ensuring precise proportions of materials in the concrete mixture. This article explores different models, including the linear model (LR), nonlinear model (NLR), and Artificial Neural Network (ANN), to predict and estimate permeability and compressive strength in porous concrete. The analysis incorporates 139 samples from various papers and experimental studies, utilizing significant parameters and variables like water-to-cement ratio, coarse aggregate content, cement content, porosity, and curing time as input variables. Statistical assessments, such as Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Scatter Index (SI), OBJ value, and coefficient of determination (R2), are employed to assess model performance. The results reveal that the ANN model outperforms other models in forecasting permeability and compressive strength of porous concrete. The SI and OBJ value of the ANN model are lower than those of all other models, indicating superior performance. The robust performance of the ANN model has significant implications for construction applications, ensuring precise material proportions and contributing to the durability of porous concrete structures. The success of the ANN model suggests avenues for refinement, including architecture adjustments and dataset expansion. These findings offer valuable insights into the ongoing efforts to optimize simulation techniques for predicting key properties of construction materials. On the other hand, the use of these models to optimize concrete mix design not only enhances efficiency but also significantly conserves raw materials and reduces energy consumption. These advancements contribute to lowering carbon emissions and promoting sustainable practices in the construction industry.
Publisher
Springer Science and Business Media LLC