Intelligent Model for Dynamic Shear Modulus and Damping Ratio of Undisturbed Marine Clay Based on Back-Propagation Neural Network

In this study, a series of resonant-column experiments were conducted on marine clays from Bohai Bay and Hangzhou Bay, China. The characteristics the dynamic shear modulus (G) damping ratio (D) these examined. It was found that G D not only vary with strain (γ), but they also have strong connection soil depth (H) (reflected by mean effective confining pressure (σm) in laboratory test conditions). With increasing H fixed γ, value gradually increases; conversely, decreases, is accompanied weakening decay or growth rate. An intelligent model based back-propagation neural network (BPNN) developed for calculation parameters. Compared existing function models, proposed avoids forward propagation data errors need human intervention regarding fitting can accurately predict at different corresponding γ. prediction accuracy universal does strictly depend number neurons hidden layer network.