LINEARITY TESTS FOR IN-DUCT ACOUSTIC ONE-PORT SOURCES

• Published in: Journal of sound and vibration Vol. 237; no. 1; pp. 45 - 65
• Main Authors: BODÉN, H., ALBERTSON, F.
• Format: Journal Article
• Language: English
• Published: London Elsevier Ltd 12-10-2000; Elsevier
• Subjects: Applied sciences; engine exhaust system; Engines and turbines; Exact sciences and technology; Internal combustion engines: gazoline engine, diesel engines, etc; load method; Mechanical engineering. Machine design; Input output; Invariant; Random noise; Sound level; Pipe system; Probabilistic approach; Input signal; Acoustic wave; Large pipe; Compressor; Acoustics; Output signal; Pipe; Sound source; Experimental study; Linear system; Piping; Plane wave; Internal combustion engine; Wave generation; System identification; Measurement method
• ISSN: 0022-460X; 1095-8568; 1095-8568
• DOI: 10.1006/jsvi.2000.3034

Acoustic one-port source data are commonly used to predict the plane wave sound generation in duct and pipe systems connected to fluid machines. The source data are usually determined experimentally, which assumes that linear time-invariant system theory can be used. Since some machines such as IC-engines and compressors generate very high sound levels in the connecting ducts or pipes it is of interest to investigate whether the assumption of linearity is justified. Linearity tests for linear system identification when both input and output signals can be measured are common in the literature. In the case when only the output signal can be measured linearity tests are not so readily found. This paper presents two different linearity coefficients for determining whether an acoustic one-port source under test is linear. Their sensitivity to random noise and their ability to detect non-linearities are investigated by simulations and measurements on several types of machines.