Large-scale models

Large-scale experiments, up to a 1:1 scale, are now possible in the new River Lab. With flow rates of up to 10 m³/s and extensive indoor and outdoor testing areas, the infrastructure allows for the correct scaling of objects such as vegetation, fish, or even humans.

It is now possible to model the full range of heterogeneous river sediments without reducing grain size, which would otherwise alter physical properties. This enables (near) 1:1 experiments with Reynolds numbers comparable to those found in natural watercourses.

Full-scale models

In full-scale models, the river section under investigation is considered in its entirety. The river course is replicated with its meanders, bank slopes, floodplains, and any hydraulic structures or installations. This accurately reproduces secondary currents, which play a crucial role in river bends, or the complex three-dimensional flows that occur when structures are circumvented or overtopped. Full-scale models of large rivers or long river sections require a corresponding amount of space. In this case, the scale often has to be chosen very small, which leads to scaling errors due to the significantly underestimated Reynolds number in the scaled model. Full-scale models are particularly indispensable when morphological developments, for example, in expansions or in rapidly changing water body geometries, are to be investigated.

Section models

In cases where the geometry of the river can be neglected due to relevant local characteristics, so-called section models are used. Here, typically a “strip” of a river is modeled in the longitudinal direction. This allows for the construction of models at larger scales. Section models are usually set up in existing laboratory flumes and are therefore easier to manufacture than full-scale models. A typical application example is determining the discharge capacity of a weir. Section models do not replicate the entire geometry of the river but only a relevant area or the most relevant parameters. Often, section models are also used for basic schematic research questions where specific local conditions do not need to be considered.

Model family

In a model family, the same experimental setup is constructed at multiple scales. Investigations at different scales make it possible to identify and quantify potential scaling errors. This improves the interpretation of the results at the natural scale. While average flow velocities are usually very well represented in scaled-down models, this is not always the case for turbulence parameters, which is partly due to the incorrectly represented Reynolds number in the scaled model. Therefore, potentially all Reynolds number-dependent processes are prone to errors, and the model family is one way to estimate these scaling errors.

Fluid-Structure Interactions

The interaction of solid bodies with the surrounding water falls into the category of fluid-structure interactions. Typical scenarios in which fluid-structure interactions are discussed include, for example, turbines, wind turbines, and sediment transport, but also the swimming behavior of aquatic organisms. As part of basic research on the movement of sediment particles in water, experiments on the motion of individual particles using 3D-PTV have been conducted. This includes experiments investigating the existing flow structures at the onset of movement of individual stones, as well as at the moment of collisions between stones. These measurements are also used to develop new numerical methods for simulating the movement of stones.

Hybrid Modeling

Measurements can often only be carried out with limited accuracy, or only at specific points. Many relevant physical quantities can also only be measured indirectly, such as the pressure in the flow field. In cases where additional results are of interest across the entire three-dimensional domain, numerical simulations are used. Such an approach is referred to as hybrid modeling. In this process, the measured data are used to calibrate the numerical simulation. Simulations then provide data across the entire flow field. This data can, in turn, be used to adjust the model experiment or to identify areas where additional measurements can provide new insights. If changes in geometry become necessary during the course of the experiments, hybrid modeling can also offer advantages. Small changes are usually easier to implement in the model experiment, while large changes are easier to test in numerical simulations.

Laser Doppler Velocimetry – LDV
Laser Doppler Velocimetry (LDV) utilizes the Doppler effect by scattering laser light off small particles carried by the flow. The velocity of the particles, and thus the flow, can be determined from the frequency shift of the scattered light compared to the original laser beam. This is a non-contact measurement method that allows for the determination of velocities at a point with high temporal resolution. Therefore, LDV is suitable not only for measuring average flow quantities but also for the turbulence analysis of flows.

Particle Image Velocimetry – PIV

Particle Image Velocimetry (PIV) is a non-contact, laser-optical measurement technique in which a plane of the flow is illuminated with a laser sheet. One (or more) synchronized cameras photograph the tracer particles moving through the illuminated section at short intervals. By using correlation-based analysis of the characteristic displacement patterns of the particles between two images, velocity fields (velocity vectors) of the flow can be determined over larger areas. Since it is an imaging technique, PIV can also be used to visualize flow fields.

3D Particle Tracking Velocimetry – 3D-PTV

3D Particle Tracking Velocimetry (3D-PTV) utilizes the principle of (Lagrangian) particle tracking in three-dimensional space (laser/light volume). By using multiple high-speed cameras from different viewpoints, the trajectories of individual tracer particles can be reconstructed with high temporal and spatial resolution. From the particle trajectories, the velocity and acceleration of the particles are derived and interpolated onto a regular grid. This allows for detailed investigation of three-dimensional flow and fluid-structure interactions and is used in research areas where a deep understanding of flow dynamics is required. Along with Particle Image Velocimetry (PIV), this imaging technique can also be used to visualize flow fields. Laser-optical methods (LDV, PIV, 3D-PTV) are sophisticated techniques for the precise measurement and analysis of flows in science and engineering.

Innovative Measuring Device Development – Color Classification in the Physical Model

To enable as uninfluenced and efficient data collection as possible in small- and large-scale model experiments, the continuous development of existing measurement technology and the creation of new, innovative, and non-contact measurement procedures are essential. For the non-contact detection of the grain composition of the bed and the grain size-dependent transport and scattering behavior of sediments, the new hydraulic laboratory at BOKU uses colored quartz sands and gravels. These are captured through high-resolution image recordings and classified into groups of defined grain size using color classification algorithms.

Innovative Measuring Device Development – Intelligent Stones

For capturing the movement characteristics of transported stones in a flowing water body, the Institute of Hydraulic Engineering and River Research (IWA) uses intelligent stones. The new, exceptionally large model scales up to the unscaled model allow for the use of state-of-the-art sensor technology in individual bedload particles. Such “intelligent” pebbles can continuously record and characterize resting phases, mobilization, and transport. The capture of acceleration, angular velocity, and the surrounding magnetic field, in combination with mathematical filtering techniques, allows for an analysis of the precise movement patterns.