This study investigates the application of Artificial Intelligence (AI) techniques to evaluate the mechanical properties of geopolymer concrete incorporating blend of Banana Peel-Ash (BPA) and Sugarcane Bagasse Ash (SCBA) at proportioning of 0-100% with sodium silicate (Na₂SiO₃) to sodium hydroxide (NaOH) ratio of 1.5-3. The research employed three AI methodologies: Artificial Neural Networks (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Gene Expression Programming (GEP) to enhance prediction accuracy for the mechanical properties of the geopolymer concrete using 104 datasets. The mix designs were optimized by varying the proportions of BPA and SCBA, Na₂SiO₃ to NaOH ratio, molarity, and aggregate to binder ratio with tests showing that an optimal combination significantly improved the concrete's mechanical properties. Experimental evaluations, including slump tests, confirmed the satisfactory workability of the mixes. The results revealed that increasing the molarity of the alkaline activator from 4M to 10M led to improved compressive and flexural strengths. Specifically, a 10M molarity with 52.5% SCBA and 47.5% BPA achieved a peak compressive strength of 33.17 MPa after 20 hours of curing, while a mix with 95% SCBA and 5% BPA at 4M had a lower compressive strength of 21.27 MPa. Similarly, the highest flexural strength of 9.95 MPa was obtained at 10M molarity, compared to 4.12 MPa at 4M. Microstructural analysis through Scanning Electron Microscopy with Energy-Dispersive X-ray Spectroscopy (SEM-EDS) revealed insights into the pore structure and elemental composition of the concrete, while Thermogravimetric Analysis (TGA) provided data on the material's thermal stability and decomposition characteristics. Performance evaluation of the AI models showed that ANFIS provided the most accurate predictions compared to ANN and GEP, reflecting its effectiveness in modeling complex relationships in geopolymer concrete properties with 0.345, 0.587, 1.409 and 0.998 was calculated for MSE, RMSE, MAE and R² respectively. This study highlights the potential of integrating AI with experimental data to optimize the formulation and performance of geopolymer concrete, advancing sustainable construction practices by effectively utilizing industrial by-products.

Projects

Currently no objects available

Currently no objects available