A Robust Industrial Procedure for Measuring Modal Sound Fields in the Development of Radial Compressor Stages

Abstract

The turbocharger is a significant noise source in large diesel engines, such as those used in container vessels. Its main noise source is the radial compressor, where improvements in silencers and turbocharger insulation have led to a considerable reduction of compressor inlet noise emission over the past few years. As a result, compressor outlet noise is now becoming increasingly significant for large engines. Recently, an in-house compressor testbed was upgraded by adding an acoustic modal measurement system (MSMS) that allows detailed investigation of modal sound fields inside the piping. This forms part of an updated compressor acoustic qualification procedure. This paper is an in-depth treatise of the characteristics of this modal measurement system. The calculation approach for the modal decomposition and a simplified alternative that assumes axial propagation, as well as relevant considerations, such as spatial resolution, averaging, and the use of multiple reference sensors, are addressed. Various measurement parameters, such as repeatability, measurement time, required temperature stability, pressure scaling, flow noise and their impact on measurement uncertainty were investigated. A successful validation of the modal sound measurement system with a well-known modal sound field at the compressor inlet is also presented. Finally, the characteristics of the modal sound fields of the compressor outlet of a typical modern turbocharger are discussed. Modal decompositions at the first two blade passing frequencies (BPFs) are presented for selected operating points (OPs). The response of total sound power levels (PWLs) to compressor speed along the operating line (OL) is examined by means of both the present and the simplified algorithm. A sensitivity analysis shows the impact of volume flow and rotational speed on the modal sound distribution.

Deep neural networks for waves assisted by the Wiener–Hopf method

“…The inversion method itself, compared to the construction of forward models, is actually straightforward. To explain this point, the compressive sensing method is used here as an example, because it has found ever-increasing applications in aerospace engineering in general [38,39] and aeroengine acoustic experiments in particular [1][2][3]. Through compressive sensing, an inversion can be achieved by solving the following linear programming [40]:…”

“…Hence, the current paper is focused on the forward modelling topic. It is also worthwhile to mention that most fan noise testing techniques [1,2] currently used in aeroengine applications belong to mode detections, for which the transfer function G would simply be Fourier expansions (that mapŝ to azimuthal modes) or Bessel functions (that mapŝ to radial modes). In contrast, the mode inversion task calls for a transfer function G that maps fan noise source to far-field measurements, which would be much more difficult to achieve (figure 3) and is therefore one of the focal points in this work.…”

“…In the training of the U-net, we deliberately choose a dataset of 10 000 items that only covers 4 modes, that is, (m, n) = (6, 1), (6,2), (7,1) and (7,2), with uniform background flows of M 0 = M 1 = 0.1. The corresponding training loss is similar to figure 10.…”

Dynamic mode decomposition of the acoustic field in radial compressors

“…Later, authors such as Raitor & Neise [6] implemented these techniques in the experimental characterization of centrifugal fans and turbocompressors. Most recently, Limacher, Banica et al investigated the sound field of centrifugal turbochargers through the least-square fitting of select blade passing frequencies to Tyler-Sofrin modes from both numerical [21] and experimental [22] points of view.…”

“…It can be seen in the aforementioned studies [6,21,22] that when noise phenomena other than BPF harmonic tones are decomposed in this way, a large number of mode orders become excited, this is, a large sum along the m and n orders is required in order to explain the observed pressure pattern at the measured cross-section.…”