Andreas Wurzinger

This paper proposes a mode-oriented finite-element model-updating workflow applicable to thin-walled duct structures with material damping. This approach replaces damped-eigenvalue analyses with point-wise time-harmonic simulations evaluated only at estimated target-mode frequencies. It pairs the simulated and measured frequency- response vectors using the frequency-domain assurance criterion. Three physically motivated objective functions are compared, optionally combining penalties based on the normalized frequency mismatch, the frequency-domain assurance criterion, and the vibration magnitude. The objective functions are minimized via surrogate-assisted global optimization, with Random Forest regressors guiding a Differential Evolution search. The workflow’s applicability is demonstrated on three scenarios of increasing complexity: S1 recovers known flange elasticity and damping in a controlled setup, S2 identifies equivalent material properties and normal-stiffness boundary conditions to emulate a more detailed bolted-flange model and a ground-fixed mounting, and S3 updates a real ventilation duct belonging to a centrifugal-fan test rig, where laser-Doppler vibrometry measurements with impact-hammer excitation provide the reference. Validation is performed by comparing the frequency-domain assurance criterion and relative mean-squared errors along a proximity band to summarize the model quality over frequency. Results yield consistent surrogate convergence and recovery of target parameters in S1, and substantial improvement over a trivial baseline in S2 and S3 across a broad frequency range. The most robust performance is obtained with an objective that combines frequency-shift and frequency-domain assurance-criterion penalties. The workflow is designed to estimate unspecified parameter values from large parameter bounds, especially for wide-band matching, models that initially deviate significantly from the reference, and situations with non-negligible damping.

Shell-like housing structures for motors and compressors can be found in everyday products. Consumers significantly evaluate acoustic emissions during the first usage of products. Unpleasant sounds may raise concerns and cause complaints to be issued. A prevention strategy is a holistic acoustic design, which includes predicting the emitted sound power as part of end-of-line testing. The hybrid experimental-simulative sound power prediction based on laser scanning vibrometry (LSV) is ideal in acoustically harsh production environments. However, conducting vibroacoustic testing with laser scanning vibrometry is time-consuming, making it difficult to fit into the production cycle time. This contribution discusses how the time-consuming sampling process can be accelerated to estimate the radiated sound power, utilizing adaptive sampling. The goal is to predict the acoustic signature and its uncertainty from surface velocity data in seconds. Fulfilling this goal will enable integration into a product assembly unit and final acoustic quality control without the need for an acoustic chamber. The Gaussian process regression based on PyTorch 2.6.0 performed 60 times faster than the preliminary reference implementation, resulting in a regression estimation time of approximately one second for each frequency bin. In combination with the Equivalent Radiated Power prediction of the sound power, a statistical measure is available, indicating how the uncertainty of a limited number of surface velocity measurement points leads to predictions of the uncertainty inside the acoustical signal. An adaptive sampling algorithm reduces the prediction uncertainty in real-time during measurement. The method enables on-the-fly error analysis in production, assessing the risk of violating agreed-upon acoustic sound power thresholds, and thus provides valuable feedback to the product design units.

Impedance tube measurements are a widespread method for determining the sound absorption coefficient (SAC) of porous materials for normal sound incidence. The measurement method is standardised in ISO 10534-2. However, the standards offer limited guidance on sample preparation and mounting. Many impedance tubes have a circular cross-section and can be mounted at various angles. Therefore, it is crucial to investigate the influence of the mounting angle on the SAC and its interaction with diameter imperfections and different thicknesses using different materials. Following the dataset FOAM 01 this work documents the creation of a new dataset (called FOAM 02) containing SAC measurements from the ISO 10534-2 two-microphone method. The following factors are considered: two different materials, four thicknesses, three diameters, and four rotation angles. For each combination of material, thickness, and diameter, three specimens are produced, resulting in 72 specimens. Each specimen is measured three times at each rotation angle, yielding 864 SAC measurements. The dataset contains the SAC measurements and one-hot encoded label vectors, and is publicly available. A cutting device is proposed to saw cylindrical specimens accurately on a band saw with variable thickness and diameter. The workshop drawings of the cutting device are available as supplementary material.

Modeling fluid-structure-acoustic interaction is a highly challenging task and currently there is a lack of available benchmark datasets. For instance, to understand the processes of human phonation, it is essential to fully understand the fluid-structure-acoustic interaction process. In this article, a synthetic human phonation model is presented for benchmarking numerical methods in the field of aerodynamics, aeroacoustics and the highly-complex fluid-structure-acoustic interaction process of the voice production. The objective is to present the wide range of experimental data available, including the model geometry, the material characteristics, quantitiesreferred to the vocal fold dynamics, fluid mechanical quantities and the acoustic field. Firstly, the experimental setup is specified. Secondly, the experimental data is described in detail, accompanied by illustrations. Having the dataset in hand, finally, it is shown how to use this data to validate a computational phonation model called simVoice successfully. In conclusion, this benchmarking dataset offers the opportunity to validate structural dynamics, aerodynamics and aeroacoustics of a highly-complex fluid-structure-acoustic interaction simulation, obtained from different mathematical formulations and numerical procedures.

The noise generation of an axial fan is mainly caused by flow-induced noise and can therefore be extracted from its aeroacoustics. To do so, a hybrid approach separating flow and acoustics is well suited due to its low Mach number. Such a computationally efficient hybrid workflow requires a robust conservative mesh-to-mesh transformation of the acoustic sources as well as a suitable mesh refinement to guarantee good convergence behavior. This contribution focuses on the mesh-to-mesh transformation, comparing two interpolation algorithms of different complexity towards the applicability to the aeroacoustic computation of an axial fan. The basic cell-centroid approach is generally suited for fine computational acoustic (CA) meshes and low phase shift, while the more complex cut-volume method generally yields better results for coarse acoustic meshes. While the cell-centroid interpolation scheme produces source artifacts inside the propagation domain, a grid study using the grid convergence index shows monotonic convergence behavior for both interpolation methods. By selection of a proper size for the source grid and source interpolation algorithm, the computational effort of the experimentally validated simulation model could be reduced by a factor 4.06.