Author(s): B. Bhattacharya; S. Sewagudde; T. Van Kessel; D. P. Solomatine
Linked Author(s):
Keywords: Keywords: suspended sediment; Dutch coast; Hybrid model; Numerical model; Data-driven model; Neural networks
Abstract: This paper presents an investigation on the performance of a numerical model when time varying sediment boundary conditions are specified instead of time invariant conditions in simulating suspended particulate matter (SPM) along the Dutch Coast. The time varying sediment boundary conditions are generated using an artificial neural network (ANN) model. The ANN model is built using measured data and data generated by a numerical model. A hybrid modelling approach is developed in using the SPM time series generated by the ANN model as a sediment boundary condition at the open boundary of a fine-grid 3D numerical model of the Dutch Coast using Delft3D. Results of this approach are compared with those when time invariant sediment concentration is specified at the open boundary. It is concluded that the sediment boundary condition provided by the ANN model provides improved simulation results and the methodology presents new horizons for developing hybrid models.
Year: 2011