Constanza Montserrat Cruz Sánchez her master’s thesis

Student: Constanza Montserrat Cruz Sánchez
Main Supervisor: Samuel Brauer, Volvo Penta
Examiner: Håkan Johansson
Opponent: Kirupakaran Yeshwanth

Abstract of thesis

The transition from a diesel engine to an electric engine represents a significant step towards achieving greater sustainability in maritime vessels. However, this conversion can expose tonal noise from the Integrated Propulsion System (IPS) component of a boat propulsion system. Therefore, the main objective of this thesis project is to develop a methodology for assessing sound transmission loss (STL) in components that incorporate absorbent materials as passive noise control measures.

The STL model was constructed using ACTRAN software. The model consisted of a two-layered system composed by wood and foam in combination with a monopole source. Simulations were conducted in a semi-anechoic chamber setup. Two mathematical approaches, namely the Johnson-Champoux-Allard (JCA) model and the Miki model, were employed to study sound propagation in porous media and evaluate their
impact on defining acoustic parameters. In order to determine the flow resistivity of any material when the acoustic properties are unknown, sound absorption and impedance theories were employed within an impedance tube model.

The results demonstrated that accurately defining the complete acoustic parameters in the JCA model is crucial for obtaining reliable results. Furthermore, for accurate prediction of flow resistivity using either sound absorption or impedance values as inputs, impedance measurements must be performed utilizing the Miki model.

The TL results exhibited a good correlation between the physical measurements and the simulations conducted in ACTRAN, using both the Miki Model and JCA model. However, it should be noted that since the Miki model has only one material parameter, is more sensitive to changes in flow resistivity compared to the JCA model. Consequently, variations in flow resistivity can have a substantial impact on the results and may result in differences between the outcomes produced by each model.