Analyzing the Effect of Gypseous Soil Stabilized with Ground Granulated Blast Furnace Slag by DBN-MFO

Gypseous soil is an emerging process in the construction environment for the application of building works presently. Increasing the strength of the gypseous soil is the difficult task due to its properties. Therefore, this work utilizing the Grounded granulated blast furnace slag (GGBS) to stabilize the gypseous soil for increase the strength. Moreover, the prediction model is developed for predict the effect of gypseous soil when stabilizing with GGBS. Enhanced Deep belief network (DBN) is employed for the prediction model where DBN is enhanced with the help of Mayfly optimization (MFO) algorithm. The proposed model is validated by utilising experimental data in terms of compressive strength, weight loss, collapsibility potential, and coefficient of permeability. The experiment is carried out using two alkali activators with different molarities, KOH and NaOH. The experimental process demonstrates that the GGBS with NaOH [12 M] has greater stability than the other mixing ratios. On the other hand, the simulation results demonstrated that the proposed model performed efficiently, with lower error values and a higher level of accuracy. The prediction model achieved 96.78% accuracy, with MSE at 0.0012, MAE at 0.032, and MAPE at 0.0096.