Digital Twin of River by 3D Modeling with Sediment Simulation for Flow Path Assessment

This study develops a Digital Twin of River by integrating sediment simulation into a previous 3D river model, addressing limitations in the previous research that lacked sediment dynamics. The proposed method enhances flow path health assessment by incorporating realistic sand particle behavior, simulating sediment transport and deposition within the river system. The simulation is developed using Blender with Python scripting, where granular physics is applied to model sand movement under varying flow conditions. The generated data is then exported to MATLAB for flow path estimation using Fourier basis (FB), Gaussian process regression (GPR), and support vector regression (SVR). The study evaluates the impact of sediment transport on river flow, demonstrating that incorporating sand improves the accuracy of flow-path predictions. Performance metrics such as the river health index (RHI) and root mean square error (RMSE) are used to compare simulations with and without sediment. The results show that sediment dynamics play a crucial role in river flow modeling, offering a more comprehensive cyber-physical system (CPS) for river monitoring. This research contributes to improved river health assessment, supporting future developments in digital twin-based environmental monitoring and flood risk management.