Towards a Centrifugal Fan Database for Off-Design Validation of CFD Simulations
B3 Analytical, Numerical & Experimental Methods II
Turbomachineries such as fans are energy conversion machines which can be found in many daily technical devices. The bandwidth of power consumption ranges from a few watts to megawatts and energy-saving requirements are getting increasingly strict with every year. Therefore, a continuous study of the flow phenomena inside such machines is necessary to identify and to understand the sources of energy losses. To analyze the flow phenomena beyond empirical findings, Computational Fluid Dynamic (CFD) simulations are nowadays being used to unlock further potentials for design optimizations and energy savings without the need for expensive experiments. While the optimal operation point typically can be described well by such simulations, the off-design conditions of turbomachineries are very challenging to predict due to unsteady effects for instance.
A validation test bench for a low-pressure centrifugal fan with a spiral casing has been built and Particle Image Velocimetry (PIV) is used to measure the flow field in several different positions and various operation points. The acquired data together with the geometry and a structured CFD mesh is provided through an open-source database and is intended to be used by researchers to validate their CFD codes and methods.
On the pathway towards the centrifugal fan database, research and meta data management along with an uncomplicated data access organization are two key building blocks. This paper highlights the concept and implementation of the data management using HDF5 as the underlying data file format. Furthermore, the workflow and tools provided for other scientists, that aim to validate their fan CFD simulations, is presented.