Session: 34-05 Harmonic balance methods

Paper Number: 125015

125015 - On the Development of an Efficient Sliding Mesh Interface for the Harmonic Balance Method, Part I: Implementation and Verification 

In recent years the Harmonic Balance method has become a popular tool in the field of turbomachinery design. While it can efficiently capture nonlinear effects in time-periodic flows, its application to complex configurations such as cavities or casing treatments requires sophisticated domain coupling methods. To address this topic we present an algorithm for the coupling of non-conforming mesh topologies across domain boundaries in a series of two papers.

In this first paper an approach based on a contact-area weighted mapping and a single communication step between adjacent surface elements is introduced. It is shown that memory layout can have a strong impact on memory usage and CPU time. To handle indexed modes, i.e., modes with the same frequency but different inter-blade phase angles, we introduce virtual passage segments on which the required spatial Discrete Fourier Transform (DFT) is performed. Using Almost Periodic Fourier Transforms (APFT) to transfer the temporal solution harmonics we show that the computational overhead can be significantly reduced when compared to a standard temporal DFT ansatz. In this context a simple, but robust algorithm is proposed to determine the temporal sampling interval and distribution of the sampling points such that a bound on the condition number of the resulting APFT matrix is satisfied. The method is verified and analysed using three academic test cases: the propagation of an entropy wave across moving interfaces in an empty duct with a rotating mean flow; the aerodynamic excitation of a simple rotor blade exposed to a time-periodic gust disturbance; and the flow in a duct with an attached cavity.

Presenting Author: Georg Geiser German Aerospace Center (DLR)

Presenting Author Biography: 2001-2008: Diploma in Mechanical Engineering at RWTH Aachen University, Aachen, Germany
2008-2014: Doctoral Degree in Aerodynamics at the Institute of Aerodynamics, RWTH Aachen University, Aachen, Germany
since 2015: Scientist at the Institute of Propulsion Technology, German Aerospace Center (DLR), Cologne, Germany
since 2022: Group Lead High Performance Computing and Artificial Intelligence at the Institute of Propulsion Technology, German Aerospace Center (DLR), Cologne, Germany

Authors:

Georg Geiser German Aerospace Center (DLR)
Felix Schwarzenthal German Aerospace Center (DLR)
Graham Ashcroft German Aerospace Center (DLR)
Maximilian Hartmann MTU Aero Engines AG
Daniel Schlüß MTU Aero Engines AG