Author(s): Clemens Cremer; Jesper Sandvig Mariegaard; Henrik Andersson; Jannik Elsasser; Faro Schafer

Linked Author(s):

Keywords: Calibration; Hydrodynamic Modeling; Optimization; Parameter Estimation; Bayesian optimization

Abstract: Calibration of hydrodynamic models is fundamental for accurate simulations, yet traditional manual methods are time-consuming and may not fully explore complex parameter spaces. While advanced optimization algorithms capable of efficiently navigating high-dimensional parameter spaces have become increasingly sophisticated, their application in hydrodynamic modeling remains limited. Such algorithms, often developed for machine learning hyperparameter tuning, offer valuable capabilities for complex parameter estimation problems. This study tests various sampling strategies for automated calibration on two distinct case studies: the Hamburg Elbe Estuary and Southern North Sea models. The optimization procedure utilizes water levels at various measurement locations, aiming to enhance calibration of key model tuning parameters while providing valuable insights through parameter sensitivity analysis. Results demonstrate improved calibration quality and generalizability while reducing manual effort and shows that Gaussian process sampling approach is superior to the alternatives

DOI: https://doi.org/10.64697/978-90-835589-7-4_41WC-P1734-cd

Year: 2025