Performance Evaluation and Prediction of Rice Husk Ash Concrete Properties Through ANN Algorithm
摘要
The growing demands for sustainable high-performance concrete (HPC) necessitates innovative solutions to reduce cement consumption while enhancing mechanical and durability properties. This study investigates the rice husk ash (RHA) effects as cement (partial) replacement in M60-grade HPC, with substitutions ranging from 5 to 30% by weight. To mitigate RHA’s high surface area, superplasticizers were incorporated to maintain workability. Comprehensive evaluations including compressive/split tensile strength, ultrasonic pulse velocity (UPV) and water absorption were conducted at duration of 7 and 28 days. Results confirm 15% RHA mix outperformed the control, achieving higher compressive strength (6.2% increase) and split tensile strength (9.4% increase), alongside lower water absorption (18% reduction) and elevated UPV values, indicating a denser microstructure. However, replacements beyond 20% led to declining strength due to reduced cement content and insufficient pozzolanic reactivity. Artificial Neural Network (ANN) algorithm was employed to get optimal RHA dosage, corroborating experimental findings. The study concludes that 15% RHA optimally enhances HPC performance, offering structural, economic, and environmental benefits by reducing cement usage and valorizing agricultural waste. These findings underscore RHA’s viability as sustainable supplementary cementitious material for high-strength applications.