Laboratory for Flow Instabilities and Dynamics

Current Projects

Data-driven modelling of physical mechanisms in combustion chambers

In the interaction with renewable energies, gas turbines originally designed for one operating point will have to exhibit significantly higher load and fuel flexibility in the future. A major challenge here is to avoid thermoacoustic pulsations in the combustion chamber for the entire operating range. By spatially resolving the dynamics of individual mechanisms of action in the combustor, the complex interaction can be studied and modeled. The resulting understanding of the instabilities allows future gas turbines to be tested for their flexibility at an early stage of development using simple measurement technology.

Thermoacoustics in industrial gas turbine combustors

Combustion instabilities are a major challenge for gas turbine and rocket engine manufacturers. Positive feedback between acoustics and heat release rate, known as thermoacoustic instability, can lead to an increase in emissions, structural damage, or even engine failure. To reduce toxic combustion byproducts, such as nitrogen oxides, modern gas turbines are operated with lean fuel-air mixtures. However, such lean premixed flames are susceptible to acoustic disturbances, so reducing pollutants poses an increased risk of thermoacoustic instability. To ensure safe operation at low emissions, a thorough analysis of the acoustic and thermoacoustic properties of the system is essential throughout the design process.

Active flow control of hydrodynamic instabilities in gas turbine combustors

In the combustion chambers of modern gas turbines, so-called premixed flames are used, which are stabilized with the help of a twisted flow. This type of reacting flow can form hydrodynamic instabilities, which also influence the combustion process. One of these instabilities, which is known as the Precessing Vortex Core (PVC), causes helical coherent vortex structures in the shear layers of the flow field. In order to investigate the direct influence of the PVC on the combustion process, the amplitude and frequency of the PVC are adjusted by means of active flow control.

Modeling and control of flow instabilities in turbulent swirling flows

Turbulent swirling flows with vortex breakdown occur in various engineering flows such as on delta wings of aircraft at high angles of attack, in water turbines at part load, or in combustors of gas turbines. Under certain conditions, vortex breakdown is accompanied by a strong flow instability called precessing vortex core (PVC). The goal of this project is to build a model-based understanding of the PVC in order to systematically investigate its influence and develop effective control measures.

Modeling of trailing edge noise on the rotor blade of a wind turbine

Wind power is a common source of renewable energy. However, noise from wind turbines is a significant problem for wider use in the onshore sector. The trailing edge noise at the turbine rotor is considered the largest source of noise. This project aims to build a model and detailed understanding of the mechanisms by which trailing edge noise is generated, in order to develop long-term control and influencing solutions to mitigate the noise.

Optimal flow control in Francis hydro turbines

Due to their flexible operational range, hydropower plants are well suited to compensate for power grid fluctuations that arise due to the volatility of renewable energies such as solar or wind power. For this purpose, hydropower plants, which are often run with Francis turbines, need to be operated at part load more frequently. Under these operating conditions, a flow instability called precessing vortex core can occur. Its presence can lead to significant losses in efficiency or even to a complete failure of the turbine. The goal of this project is, therefore, to facilitate a safer operation of the part load regime by attenuating or suppressing the flow instability through flow control.

Development of gas turbine combustors through artificial intelligence and machine learning

The goal of the project is to accelerate the design process and discover new, unconventional configurations and modes of operation for gas turbines through the use of artificial intelligence. Machine data analysis and the adaptation of models by AI have experienced a surge in development in recent years, mainly due to increasingly available computing capacity, and are finding more and more applications in a wide range of scientific fields. In order to quickly adapt the design process of gas turbine combustors to new requirements, they will be used to perform automated experimental and numerical investigations and efficiently evaluate data. In this way, stable operating points with minimal pollutant emissions are to be found efficiently.

Linear stability and resolvent analysis for modeling wave packets in broadband turbulent flows

The far field of round turbulent jets is characterized by the appearance of broadband turbulent structures. The focus of this project is the identification of the occurring coherent structures and their modeling by linear stability analysis and resolvent analysis. The experimental investigation of the structures is based on time-resolved stereoscopic PIV measurements, which allows the identification of coherent structures in frequency domain. The measured data additionally serves as a basis for the modeling approaches and the validation. In both applied modeling methods, the time-averaged flow field is used as a basis to predict the turbulent structures.

Linear and nonlinear Dynamics of two-phase flows

Two-phase flows describe the interaction of fluids that have different aggregate states (e.g. liquid and gaseous) and occur in a variety of technical applications. These include, for example, fluid mixing or atomization by means of nozzles or the separation of fluids in industrial cyclones. In contrast to single-phase flows, these flows sometimes exhibit significantly more complex dynamics. This research project focuses on the development of methods for modeling these dynamics and the description of the dynamics itself, for selected flows.

Linearized mean-field models for predictions of flame instabilities in turbulent flames

Combustion instabilities, i.e. temporal fluctuations in heat release, are the cause of many problems in the operation of gas turbines. They lead to increased pollutant emissions, and increased noise emissions. The latter problem is particularly crucial for aero gas turbines. In the worst case, a resonance effect occurs between the combustion instabilities and the acoustic chamber mode. In this case, one speaks of thermoacoustic instabilities, which can even lead to the destruction of the machine. These effects occur - in comparison to the combustion of conventional fuels - especially in the combustion of green hydrogen. In order to make them controllable, we are developing numerical methods in this project which are based on the linearized flow equations and help to better understand the flame dynamics.

Experimental Fatigue and Extreme Load Control employing Trailing Edge Flaps

Experimental analysis of active flow control to alleviate fluctuating loads on wind turbine blades are conducted on the Berlin Research Turbine. The Turbine is equipped with a large number of sensor and actuators to drive trailing edge flaps. Within the large wind tunnel (GroWiKa) reproducible (disturbed) inflow conditions are created that the turbine is exposed to. In the current project trailing edge flaps are employed to alleviate fatigue and extreme loads.

The FELiCS Project

The software package FELiCS (Finite Element Linearized Combustion Solver) is the focus of the development of linearized methods at the Laboratory for Flow Instabilities and Dynamics. FELiCS is a CFD tool that addresses the governing equations of the flow linearized around the temporal mean state. These equations describe the dynamics of the flow. Despite its great potential for controlling the dynamics in technical flows, its application, so far, has been mostly limited to academic configurations. The goal of the FELiCS project is to extend the scope of linearized methods to engineering applications with the aim to understand and control fluid dynamics. The challenge is to take into account the multitude of physical effects that play a role in engineering applications. These include, for example, turbulence, mass transport, heat transport, chemical reactions, acoustics and many more. To properly account for such multi-physics in the linearized set of equations, a well-founded software framework is needed that can be extended step-by-step, similar to conventional CFD programs that account for the non-linear equations. FELiCS is our answer to this challenge.

Completed FLOW projects

Pulse detonation combustors in gas turbines

The use of gas turbines is essential for the future stable energy supply as well as propulsion technology in aviation. One possible concept for a comparatively drastic increase in efficiency starts with the combustion process, which represents the most fundamental step in the energy conversion of a gas turbine. Within the framework of the Collaborative Research Center at the Technical University of Berlin (SFB1029), the classical constant-pressure combustion is to be replaced by an approximate constant-volume combustion process. Among other things, this can be realized by using pulsed detonation combustion (PDC).

More about the project