Yannick Marschall

Last Name

Marschall

First Name

Yannick

ORCID

0000-0002-7192-4468

Organisational unit

03820 - Boes, Robert / Boes, Robert

Show all metadata

Search Results

Publications 1 - 10 of 10

Flow around an isolated boulder-like obstacle: effects of modeling approach and Reynolds number
Marschall, Yannick; Constantinescu, George; Boes, Robert; et al. (2024)
Proceedings of the 10th International Symposium on Hydraulic Structures (ISHS 2024)
We conducted unsteady Reynolds-Averaged Navier Stokes (URANS) and detached eddy simulations (DES) to simulate the flow around a boulder-like obstacle placed at the bottom of an open channel at laboratory scale (obstacle width B=7 cm) and at a larger scale (B=21 cm), while keeping the channel Froude number constant. The obstacle shape was generated from a three- dimensional scan of a stone. The study discusses the effects of the turbulence modelling approach (URANS vs. DES) and scale effects. An increase of the drag coefficient of around 10-15% was observed for the simulations conducted at larger scale. Part of this difference can be attributed to differences in the fully developed velocity profiles in the approach flow which are a function of the Reynolds number. Maintaining the boulder size, the drag coefficient predicted by URANS is about 15% smaller than the DES value. URANS overestimates the wake recirculation region leading to a slightly higher pressure acting on the downstream side of the boulder. The results of this investigation show that URANS methods, which are widely used for practical applications in hydraulic engineering, can predict reasonably well the mean flow field and the quantities of engineering interest. Computationally more expensive methods like DES, that resolve a large part of the energetically important eddies in the flow, provide more accurate predictions of the mean quantities and allow investigating the flow physics based on the simulated dynamics of the large-scale coherent structures.
Tranquilizing Racks in Desanding Facilities: State of the Art and Field Study on Flow Field Improvement by Rack Modification
Kastinger, Maximilian; Felix, David; Marschall, Yannick; et al. (2024)
Journal of Hydraulic Engineering
Desanding facilities (DFs) limit particle sizes and reduce the suspended sediment load in hydropower plants. Multiple successive tranquilizing racks (TRs) at settling chamber inlets are a cost-efficient structural measure to homogenize the flow and reduce local velocities and turbulence in the chambers. We summarized the state of the art of TRs and conducted detailed flow velocity measurements in the settling chambers of a DF with TRs. In the first campaign, a pronounced flow concentration beneath the vertical bars of the TRs was found, resulting in flow velocities along the invert of up to 3.5 times the average cross-sectional velocity. Because this impairs the trapping efficiency, the TR bars were extended downward to achieve a more even distribution of the flow resistance over the entire flow section. A subsequent measurement campaign confirmed a more homogeneous flow field and a considerable reduction of the turbulence intensities and turbulent kinetic energy, which were the lowest compared to six other DFs. The results highlight the potential to modify existing TRs at other DFs with inhomogeneous flow fields and the practical importance of TR design for new DFs.
On the Role of Free-Surface Treatment for Simulating Flow Past Submerged Obstacles
Marschall, Yannick; Constantinescu, George; Boes, Robert; et al. (2023)
Proceedings of the 40th IAHR World Congress
Numerical simulations are a powerful tool to gain deep insight into the hydrodynamic processes which occur around submerged obstacles like boulders. Over the last decades, different simplified approaches were developed to reduce the computational cost of such simulations. In the popular rigid-lid approach, the water surface is treated as a rigid boundary which results in a reduction of the complexity and a decrease of the computational domain. In this study, we conducted eddy resolving simulations around submerged boulder-like ellipsoids to compare the performance of the rigid-lid and the free surface approach with regard to the flow field and drag and lift forces. Simulations have been conducted for three levels of submergence, from slightly emerged to deeply submerged ellipsoids. We observed good agreement of the flow field and flow parameters between both approaches for the deeply submerged test case. However, for a case with slight submergence, an underestimation of the drag coefficient of more than 25% could be observed with the rigid-lid approach. For even lower submergence levels, different flow structures developed leading to a significant underestimation of the drag forces and overestimation of the vertical uplifting forces. While the results of the deeply submerged case confirm the validity and applicability of the rigid-lid approach, special care has to be taken for lower submergence.
3D numerical simulation to predict flow on unstructured block ramps
Marschall, Yannick; Boes, Robert; Vetsch, David F.; et al. (2025)
River Flow 2024: Proceedings of the 12th International Conference on Fluvial Hydraulics, Liverpool, UK, 2nd- 6th September, 2024
We conducted eddy resolving three dimensional numerical simulations over an unstructured block ramp with a slope of 2% and validated the results using data from laboratory experiments. The idealized obstacles on the ramp were semi-spheres placed in a regular arrangement. For two different discharges, Q = 10 l/s and 20 l/s, the flow field was compared. For the lower discharge, we observed a strongly undulating water surface with flow depth varying between 1.3H and 2.3H, where H is the protruding height of the submerged obstacle. Furthermore, drag forces on the obstacles were determined from the simulation results. For the higher discharge and thus larger submergence, the water depth varies significantly less, i.e. between 2.2H and 2.5H. In our simulations, the undulating flow behavior leads to drag forces of similar strength as the one observed in the simulation with a higher discharge, despite lower flow velocities. We expect that an irregular arrangement of the obstacles will disturb the development of these flow conditions and eventually lead to lower drag forces acting on the individual obstacles.
Fish protection and guidance at hydropower intakes with novel f-curved bar rack-bypass systems: lessons learnt from two Swiss case studies
Albayrak, Ismail; Marschall, Yannick; Leuch, Claudia; et al. (2024)
Simulation of flow around a wall-mounted semi-ellipsoid-comparison of URANS and DES modelling approaches
Marschall, Yannick; Constantinescu, George; Boes, Robert; et al. (2022)
7th IAHR Europe Congress: Innovative Water Management in a Changing Climate. Abstract Book
Optimierung der Anströmbedingungen eines Fischleitrechens beim KW Herrentöbeli mittels 3D-numerischer Simulation
Marschall, Yannick; Leuch, Claudia; Albayrak, Ismail; et al. (2023)
Wasserbau - krisenfest und zukunftsweisend. Beiträge zum 21. Wasserbau-Symposium der Wasserbauinstitute TU München. Tagungsband 2
Flow Characterization in Macro-Rough Channels Using Detached-Eddy Simulations
Marschall, Yannick (2025)
VAW-Mitteilung
A Curved‐Bar Rack‐Bypass System With Innovative Foil Shaped Bars: Hydraulics, Operational Aspects, and Swimming Behavior of Downstream Moving Brown Trout
Yang, Fan; Moldenhauer-Roth, Anita; Marschall, Yannick; et al. (2025)
Water Resources Research
Curved bar rack ‐ bypass systems (CBR‐BS) are a promising technology for diverting downstream moving fish away from hydropower intakes. A CBR functions as a mechanical behavioral barrier by creating high velocity and pressure gradients along its bars, guiding approaching fish that exhibit avoidance reactions towards the bypass. The original CBR design features a bar shape with spacings that narrow from the upstream to the downstream tip, which raises concerns about clogging due to foliage and floating debris. To address this issue, new foil‐shaped bars (f‐CBR) were developed with similar hydraulics while maintaining a constant bar spacing along the rack. To test its performance, a 1:1 physical model of a f‐CBR with a bar spacing of 25 mm was installed in an etho‐hydraulic flume. Live‐fish tests using brown trout were conducted at approach flow velocities of Uo = 0.15, 0.3, 0.6 m/s, with bypass‐to‐approach flow velocity ratios set at VR = Uby/ Uo = 1.1 and 1.2. Additional tests with leaves and driftwood were performed to evaluate clogging probability of the f‐CBR. Flow fields were also numerically simulated for the tested flow conditions and linked to the fish behavior. Over all tested setups, fish guidance efficiencies between 47% and 66% were observed, while fish protection efficiencies ranged from 65% to 86%. These findings are compared with those from other types of behavioral fish guidance racks tested in both laboratory and field settings. Finally, potential measures for the optimal design of a f‐CBR‐BS are presented.
Flow Characterization in Macro-Rough Channels Using Detached-Eddy Simulations
Marschall, Yannick (2025)
Low-gradient mountain rivers (Slope S < 4%) are often located in the densely populated main valleys of alpine regions with important infrastructure. These rivers can be subject to large fluctuations in discharge and hence a wide range of flow conditions. For example, meteorological events cause rapidly increasing discharges, whereas seasonal effects lead to low- and high-flow conditions over longer periods. Therefore, it is of great interest to understand the flow conditions in such rivers to prevent bed and bank failure, and hence damage to nearby infrastructure. In recent years, a popular and nature-based measure to increase bed stability is the arrangement of macro-roughness elements, i.e., large boulders, on the river bed. This increases the overall roughness of the bed and thereby increases the flow resistance. Such structures are also called unstructured block ramps. However, the flow characteristics in such macro-rough channels, i.e., channels containing macro-roughness elements, are very complex and not fully understood. In this study, high-resolution 3D numerical simulations (detached-eddy simulations) are used to investigate the flow characteristics in channels with macro-rough beds. The channels exhibit typical boulder densities of 15% to 25% and a water depth similar to the protruding height of the large boulders. In a first step, common simplifications for numerical simulations are evaluated. While the Froude similarity approach (i.e., reducing the size of the computational domain while keeping the Froude number constant) gives a satisfying agreement between results of different scales, the Reynolds-Averaged Navier-Stokes simulation method and the rigid-lid approach (i.e, representing the water surface by a fixed plate) result in a deviation of the flow field and the forces acting on the obstacle that are not acceptable in the context of this project. Subsequently, detached-eddy simulations with a free surface approach are used and successfully validated using laboratory experiments of the flow in a macro-rough channel. In a two-step approach, the complexity of the model increases. In a first step, the flow field around isolated obstacles with different shapes (ellipsoidal and spherical) and for different relative submergence and bed roughness are investigated. The simulations reveal that ellipsoidal obstacles are generally favorable as they experience lower forces, which is advantageous in terms of stability. In addition, they generate lower turbulence quantities in the flow, which is preferable for aquatic species. Finally, ellipsoidal obstacles can be rotated around their vertical axis to create a more heterogeneous flow field compared to more symmetric obstacles such as spheres. In a second step, a ramp with a slope of 2% and obstacles placed in a staggered arrangement is investigated. Various parameters such as flow conditions, obstacle shape, placement density and bed roughness are evaluated. It is observed that the velocity profile below the top of the obstacles can be well described with a modified exponential flow profile. Above the top of the obstacles, a logarithmic flow profile cannot develop if the water depth is less than approximately twice the obstacle height. Regardless of the relative submergence, the highest turbulence quantities are found in the region just downstream of the obstacles and approximately in the upper half of the obstacles height. The obstacles are also shown to contribute significantly to the total flow resistance. Thus, more than 80% of the total forces and stresses are taken up by the obstacles and only a small part is exhibited by the bed. Hence, the obstacles contribute significantly to the overall stability. For hydraulic structures such as unstructured block ramps, it is shown that the orientation of the obstacles (angle between major axis and flow direction) should be uniformly distributed over the ramp, i.e., rows in which all obstacles are oriented the same way should be avoided. In addition, a placement density of 25% shows advantageous characteristics in terms of stability and ecology. In the final steps of a design process, realistically shaped obstacles, i.e., boulders, should be used, as the obstacle shape has a strong influence on the drag. The use of e.g., ellipsoids leads to inaccurate estimations of the stability and an underestimation of the flow resistance. Further recommendations are provided for the design process and the design itself of unstructured block ramps.

Publications 1 - 10 of 10

Yannick Marschall

Last Name

First Name

ORCID

Organisational unit

Filters

Search Results