1Research Scholar, Department of Civil Engineering, Zakir Hussain College of Engineering and Technology, Aligarh Muslim University, Aligarh (Uttar Pradesh)
2Professor, Department of Civil Engineering, Zakir Hussain College of Engineering and Technology, Aligarh Muslim University, Aligarh
Online published on 5 December, 2025.
Generally, bed sills are provided at the stream bed to reduce bed degradation. The local scouring around bed sills results-due to increased bed shear stress caused by disturbance generated by overflowing water. Researchers carried out-physical modelling and developed regression-based models to scour depth prediction. Various researchers employed-soft-computing techniques and found that these techniques were more effective in modelling scour than regression analysis. This study applies hybrid ANFIS techniques to predict relative scour depth at bed sills. Biogeography-based-optimization (BBO), cultural algorithm (CA), particle swarm optimization (PSO), genetic algorithm (GA), and invasive weed optimization (IWO) were used to optimize of hybrid ANFIS models. All the hybrid models showed better prediction accuracy than the traditional regression model. ANFIS-IWO showed superior prediction accuracy, followed by ANFIS-BBO, ANFIS-CA, ANFIS-PSO, and ANFIS-GA. The results indicated that all models developed in this study achieved higher prediction accuracy compared to the regression-based models available in literature. Sensitivity analysis was also-carried out to determine the influence of predictor variables. The findings of the study suggest that civil engineers can-accurately forecast the scour depth at bed sills using the ANFIS-IWO model.
Biogeography-based optimization (BBO), Cultural algorithm (CA), Genetic algorithm (GA), Hybrid ANFIS model, Invasive weed optimization (IWO), Particle swarm optimization (PSO)
