A NOVEL AI TECHNIQUE FOR FORECASTING THE COMPRESSIBLE STRENGTH OF PAVEMENT WITH ASH PARTICLES

Abstract

Understanding cementitious composites' mechanical properties, particularly compressive strength (CS), is crucial for safety, with AI approaches being particularly useful for forecasting CS with ash particles.In this study, we proposed the ensemble honeybee mating optimized dynamic artificial neural network (EHBMO-DANN) for forecasting the CS of pavement. 7 characteristic Fly ash (FA), coarse aggregate, cement, super plasticizer, fine aggregate, days, and water were used as inputs in the experimental technique to forecast the output or the CS variable. Measured at 7 and 28 days, the CS of the concrete specimens without FA of the same age is compared to those with replacement rates of 15%, 30%, and 45%. The experimental evaluation metrics using the high degree of predictive accuracy are shown by its high Coefficient determination R2, MAE, MSE, and RMSE.The construction and engineering sectors may benefit greatly from our research on the CS of pavement comprising ash particles.

Keywords

• compressive strength (cs)
• pavement
• concrete
• ensemble honeybee mating
• optimized dynamic
• artificial neural network
• (ehbmo-dann)
• ash particles
• cement