The hydraulic behaviour of filling jets inside navigation locks plays a fundamental role in the redistribution of momentum within the lock chamber and in the generation of hydrodynamic forces acting on vessels during filling operations. Reduced-order numerical tools, such as VulSluis, are commonly employed in engineering practice to estimate flow conditions and ship forces while maintaining limited computational requirements. The accuracy of these models nevertheless depends strongly on the capability of the adopted jet schematization to reproduce the actual hydraulic behaviour of the filling jet. This study investigated the downstream evolution of filling jets inside navigation locks using a combined numerical and experimental approach. This study focused on the spreading behaviour of the jet, progressive downward movement of the velocity core toward the bottom boundary, and influence of the vessel on the downstream velocity distribution. The work combines analyses performed with the MATLAB-based reduced-order model VulSluis and Particle Tracking Velocimetry measurements conducted on a physical scale model at Flanders Hydraulics using the MiniShaker system and DaVis processing environment. Experimental velocity profiles were reconstructed for different hydraulic and geometrical configurations, with and without the presence of a vessel inside the lock chamber. A comparison between the experimental measurements and the original VulSluis implementation revealed several systematic discrepancies. The original schematization tends to reconstruct velocity profiles that are excessively symmetric and vertically diffused compared to the measured flow fields. The downstream sinking behaviour of the jet can be reproduced using a manually imposed jet inclination angle, whereas the influence of the vessel is not directly included in the jet reconstruction. Based on the experimental observations, a new jet parametrization was proposed. The modified formulation combines a spreading law with a new exponential parametrization introduced in this research to reproduce the downward movement of the jet. The velocity profiles were reconstructed using Gaussian distributions calibrated directly against experimental measurements. Additional ship-dependent parameters were introduced to include the influence of the vessel position, particularly the bow location, on the jet trajectory. The modified parametrization showed better agreement with the experimental results. Improvements were observed in the reconstruction of the jet concentration, downstream sinking behaviour and wall-jet development. Simultaneously, the model preserves the reduced-order computational philosophy of VulSluis. Therefore, the proposed methodology represents the first step toward a more physically realistic representation of filling jets in engineering-oriented lock-filling models.
All data in the Integrated Marine Information System (IMIS) is subject to the VLIZ privacy policy
