Inverse Analysis of Experimental Scale Turbidity Currents Using Deep Learning Neural Networks

Abstract

This study performed inverse analysis on turbidity currents using a machine learning method. Flume experiments were conducted to verify the method. Turbidite, the deposit of turbidity current, is an active area of study because it is closely related to the exploration of petroleum resources. Since turbidites are often deposited as a result of tsunami events, the understanding of turbidity currents can also contribute to geohazard prevention. The inverse analysis method proposed in this study can help enhance our understanding of the flow properties of turbidity currents.