Regularized Machine Learning for System Identification of Ship Free-Running Manoeuvres from CFD-Based Synthetic Data: A Comparative Study
arXiv:2606.17121v1 Announce Type: cross Abstract: This study investigates supervised machine learning techniques for identifying ship hydrodynamic coefficients from CFD-generated data from free-running simulations. Specifically, ordinary least squares and regularized regression methods are applied to Abkowitz-type manoeuvring models. Training and validation datasets are derived from URANS simulations of zig-zag and turning circle manoeuvres, which are validated against experimental benchmark data. The analysis evaluates the effects of coefficient set size, minimum training length required for
The increasing availability of high-fidelity CFD data and advancements in machine learning techniques are converging to enable more sophisticated system identification methods for complex physical systems like ship hydrodynamics.
This development could significantly improve the accuracy and efficiency of ship design, simulation, and control systems, potentially leading to safer and more fuel-efficient maritime operations.
Traditional reliance on extensively empirical or purely theoretical models for hydrodynamic coefficients can be augmented or potentially replaced by data-driven machine learning approaches, reducing development time and costs.
- · Naval architects
- · Ship manufacturers
- · Maritime simulation software developers
- · AI/ML researchers in fluid dynamics
- · Traditional empirical modeling firms (if slow to adapt)
More accurate predictive models for ship performance under various conditions become available.
Reduced need for extensive physical towing tank experiments in early design phases, accelerating design cycles.
Enhanced autonomy and decision-making capabilities for uncrewed surface vehicles (USVs) due to superior understanding of hydrodynamic responses.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.LG