Source spectrum model for merchant ship radiated noise in the Yellow Sea of China. (15th November 2020)

• Record Type: Journal Article
• Authors: Jiang, Pengfei; Lin, Jianheng; Sun, Junping; Yi, Xuejuan; Shan, Yuanchun
• Is Part Of: Ocean engineering. Volume 216(2020)
• Publication Date: 2020-11-15
• Subjects: Merchant ship -- Source spectrum model -- Radiated noise -- Yellow Sea of China; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2020.107607
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: In past decades, the technologies of ship-building and engine-manufacturing have developed quickly, which may alter underwater radiated noise (URN) source levels. The URN from many merchant ships was recorded by self-sustained hydrophones mounted on the seabed near the ship lanes in the Yellow Sea of China. Twenty-seven samples of the recording are selected based on ISO and ANSI standards. The sources were assumed obeying Gaussian distribution in depth direction, and the ship source spectra were computed as a logarithmic average of source spectra calculated for different positions of ships to improve the fault tolerance of source position information. A key conclusion is that the estimated broadband ship noise source levels between 20 and 1000 Hz by RANDI-3 model are generally much greater than the measured results for ships whose length is larger than 200 m, though they are close for ships with length being less than 200 m. Furthermore, the data samples were divided into two categories to analyze. With each category, the 20–1000 Hz broadband noise of modern merchant ships is not highly correlated to ship length but closely related to ship speed. Finally, noise source spectrum models of the two ship categories were established, which express ship noise source spectra as a function of frequency, ship length and ship speed. The models can represent experimental results properly. Highlights: The ship samples were divided into two categories by length and their noise source spectrum models were established respectively. The fitting results are better than those estimated by RANDI-3 model.