top of page

Mitigating ship noise using bubble injection

This news article is part of Issue 2 of our annual newsletter, which describes our research in 2022. To read more articles describing our research progress from our second year, read the newsletter in full here.

Our colleagues at MARIN have been performing model tests to evaluate the effectiveness of two air injection systems for reducing ship underwater radiated noise.

The ‘Masker’ system aims to mitigate machinery room noise by injecting a bubble curtain over the keel and sides of the hull, thereby isolating the ship hull from the surrounding water. The ‘Prairie’ system aims to mitigate the noise of propeller cavitation by injecting air into the flow directly upstream of the propeller. Introducing air into the cavity adds compressibility to the water vapour, which weakens the cavity collapse. Both systems have been applied to naval ships for decades, but there is a lack of quantitative information concerning their effectiveness in open literature.

SATURN is examining the potential of these solutions to reduce underwater noise in the context of large commercial vessels like container ships or ferries. One advantage of these technological solutions is that they can be applied to new build vessels or retrofitted to existing ships. For MARIN’s experiments, a 94 m coastal tanker, known as the Streamline tanker, is being used for the model tests.

View of the 7.8 metre long ship model hull with steel midsection.

The Masker System

You might remember that in 2021 we conducted exploratory model tests on the effect of bubble curtains (see this article from our first newsletter). Those tests focused on whether air injected using porous hoses formed an air layer beneath the hull (aka the ‘Masker’ system). The outcome of those tests was positive, and that system has now been evaluated in terms of its effectiveness in reducing the underwater radiated noise from machinery inside the ship.

To represent a real ship structure, a ship model with a metal mid section was used. A shaker introduced broadband vibrations in the structure, which created underwater radiated sound, thus mimicking the effect of a ship’s engine room. The sound was measured at various ship speeds and air flow rates, including a baseline case with the injection system switched off.

The ‘insertion loss’ (noise reduction) due to the bubble layer is determined by taking the difference between the measured sound spectra of the tests with air and those without air. An initial analysis of the results revealed an insertion loss of up to about 22 dB! However, the insertion loss was found to be negative at certain low frequencies; i.e., the air injection system led to an increase in noise levels. This could be due to bubble resonance. The insertion loss depends on the void fraction (i.e. amount and size of bubbles) which is a function of both air flow rate and ship speed, and was measured at various positions over the hull midsection. These results are currently being analysed so that the relation between insertion loss and void fraction (as well as flow rate and ship speed) can be established.

Above Left: A snapshot of air bubble layer under the ship model hull, injected through porous hoses (black vertical strip on left hand side). The model is sailing from right to left in the left photograph. Above Right: Snapshot of air injection into the propeller disc using Prairie system.

The ‘Prairie-like' System

For tests of the Prairie-like system, another model of the Streamline tanker was manufactured. A duct was added just upstream of the propeller, fitted with needles that injected small bubbles into the flow. The noise generated by the cavitating propeller was measured both without air injection and with air injection at various air flow rates.

A preliminary analysis of the performance of the system showed an insertion loss of up to 7 dB, located in a frequency range typically containing high cavitation noise levels (about 80 Hz on full scale). We note that the increase in noise levels (negative insertion loss) at lower frequencies (up to about 30 Hz on full scale) has also been reported in the literature. The cavitation dynamics and interaction with the air was also observed with high-speed video cameras (see video below), and a special optical technique has been used to determine the size distribution of the injected bubbles.

Cavitation dynamics and interaction with the air captured with high-speed video cameras at MARIN.

It is concluded that good and complete datasets have been acquired to evaluate the use of air bubbles to mitigate machinery noise and propeller cavitation noise. Preliminary results are promising. The data is presently being analysed in more detail, while activities to model the mitigation mechanisms will start soon. This gives the shipping industry more options to reduce the impact of shipping noise on sea life.

Left: Preliminary results for Prairie system: insertion loss in decidecade bands computed for one air flow rate and propeller operating condition for Prairie system. Both model and full scale frequencies are shown in the figure. Right: Preliminary results for Masker system: insertion loss in decidecade bands determined for one air flow rate and model speed. The frequency axis relates to model scale, but it is expected that the frequencies do not change significantly for full scale.

bottom of page