UBC researchers studying AI to reduce sea noise for whales
Posted December 15, 2022 12:06 pm.
Last Updated December 15, 2022 1:06 pm.
Researchers at the University of British Columbia (UBC) are looking at artificial intelligence (AI) as a way to reduce chronic ship noise that can harm whales and other marine mammals.
Sea noise from ships can harm the hearing and feeding patterns of whales, dolphins and porpoises, researchers say. Associate Professor at UBC’s department of mechanical engineering Dr. Rajeev Jaiman says the harmful sea noise comes from the popping of bubbles created by the rotation of ships’ propellers.
“Propeller noise can hit 170 decibels, the equivalent of a jet engine or a rocket lift-off,” he said, adding it can be heard up to 100 kilometres away.
Jaiman and fellow researchers are looking at ways to change the way propellers work so they can produce less noise when around animals. That includes injecting fluid to control the movement of machinery and introducing serrated edges to the blades to change the flow patterns that cause the noise.
A 3D printed metallic strip, similar to the kind that the group is considering for making ship components from. (Devan Power/UBC Faculty of Applied Science)
Co-lead of the project and naval architect Dr. Jasmin Jelovica says some of these changes could reduce the weight of the machinery by up to 50 per cent.
“Advanced structures could be better noise barriers. They can be stronger and have other benefits as well,” he said.
But any of those changes, Jelovica notes, take time to analyze, given the impacts on safety and other aspects.
Read More: 2022 sees record number of humpback whales in B.C.
The researcher says AI and machine learning can help cut down on some of the human analysis needed when figuring out the impacts of the changing technology.
“Modelling a ship using traditional computational methods takes time. Modelling even a tiny proposed structural change can take weeks. With AI, that work can be done in a matter of seconds,” Jelovica said.
The researchers have received federal funding to build a “machine learning-based noise-prediction toolkit” that could lead to ships adjusting their noise levels based on if there are whales or other mammals nearby.
Jaiman notes that the technology would not only help the health of the animals in the sea but also cut down on carbon emissions.
Related Video:
Additionally, the researchers say the marine industry is receptive to new innovations like these.
“They recognize the need to change and to become more sustainable and environmentally friendly,” said Jelovica.
