Background: Axial fans in heat pumps and air conditioning systems are often exposed to disturbed inflow caused by upstream components such as heat exchangers. This disturbed inflow increases the sound radiation of the fan, an effect known as an “installation effect.” The Institute of Fluid Mechanics (LSTM) has developed an Active Turbulence Grid (ATG) that can generate a wide range of defined disturbed flow conditions in the laboratory. In this project, AI methods (neural networks) will be used to link the ATG control parameters to the resulting flow fields. The goal is to conduct extensive measurements of various operating modes to generate different turbulence fields and, using AI and the training sets, to be able to predict which operating parameters are necessary to generate specific turbulence fields in future.
Specific tasks:
- Support of experimental investigations at the axial fan test bench (ATG, hot-wire and
total-pressure measurements)
- Automation and control of measurements using LabVIEW
- Processing and analysis of measurement data in MATLAB
- Applying AI/machine learning methods to correlate flow field and ATG parameters
- Documentation of experimental setups and results
Requirements:
- Good working knowledge of LabVIEW and MATLAB
- Interest in / basic knowledge of AI and machine learning methods (e.g. neural networks)
- Interest in experimental fluid mechanics and turbulence
- Independent, reliable working style
Starting date: immediately
Advisor:
Alper Akardere
Research associates
Contact
Dr.-Ing. Felix Czwielong
Research associates