• 2025-04-25
• A.O. Abioye
• Volume 1, Issue 1

Abstract

This study presents Support Vector Regression (SVR) model for predicting the Ultimate Load Capacity at Failure on Rice Husk Ash (RHA) – Cement Blends Reinforced Concrete Beams. The experimental dataset comprising 120 beam specimens for both flexural and shear were used to train and validate the SVR model. The concrete beam specimens were varied at 2.5% RHA as a partial replacement by weight. Concrete cubes were cast, cured and tested at 7, 14, 21, 28, 60, 90, and 180 days and the reinforced concrete beams were cast, cured and tested at 28, 60, 90, and 180 days. RHA optimum calcination at 600 °C for 90 minutes was suitable for the concrete. The cube has highest compressive strength of 26.79 N/mm2 at 5% cement replacement with RHA on 28-day curing. The highest flexural and shear ultimate load at failure of the reinforced concrete beams at 28 days was at 5% replacement of cement with RHA. The SVR models demonstrated excellent predictive performance achieving Mean Absolute Error of 1.3272% and 1.3926, and a coefficient of determination (R2) of 0.9867 and 0.9888 for both flexural and shear beam specimens respectively. The model provides a rapid forecast and helps construction engineers to avoid mixes