Large-scale experimental and ANN modeling for dynamic interaction between vibrating and statically loaded foundations on geogrid-reinforced soil beds

The present investigation includes experimental and ANN-based intelligent modeling to explore the dynamic interference effect of closely positioned vibrating foundations placed on unreinforced and geogrid-reinforced soil beds. Large-scale field block vibration tests are conducted on isolated and interacting block footings placed on prepared foundation beds at IIT Kanpur, India. The dynamic interaction of various combinations of two-footing assemblies is examined where one footing (active footing) is excited with dynamic loadings, and the other (passive footing) carries static loadings. The tests involve three eccentric force settings for four distinct footing combinations at different clear spacings and reinforcement conditions. The responses of both footings are recorded at different loading frequencies. The interaction effect is presented in terms of the transmission ratio plotted against the frequency ratio. Additionally, an Artificial Neural Network (ANN) model is developed using the recorded field datasets to anticipate the dynamic interference effect. The predicted outcomes of the ANN model demonstrate promising agreement with the experimental findings reported in the literature, indicating the reliability and robustness of the intelligent model.