A novel hybrid superposition method for predicting ship seismic wave in shallow sea. (1st September 2022)

• Record Type: Journal Article
• Authors: Wu, Shaowei; Xiang, Yang; Qin, Tao
• Is Part Of: Ocean engineering. Volume 259(2022)
• Publication Date: 2022-09-01
• Subjects: Shallow sea -- Elastic seabed -- Ship seismic wave -- Low frequency noise -- Hybrid superposition; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2022.111948
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: A hybrid superposition method (HSM) is developed for predicting seismic wave in elastic seabed excited by low frequency noise from ship submerged in shallow sea. In the scheme of HSM, an array of distributed equivalent sources interior to a radiator are employed to represent this radiator based on wave superposition principle. The equivalent sources that satisfy the boundary conditions of shallow sea are constructed by making inverse Hanker transform to the depth-dependent Green's function corresponding to displacement potential by aid of wavenumber integration algorithm. The strengths of equivalent sources are determined by matching normal velocities on the boundary surface of radiator and then modified for ship seismic wave computation. The ship seismic wave is conveniently calculated using the superposition of the fields from the equivalent sources. It is the key feature of HSM that truncating and discretizing the field space are not required for the prediction, which reduces original problem by one dimension, while accurate results are achieved. The performance of the present method including accuracy, stability, and convergence is studied numerically. Simulations indicate that this novel method is outstanding and competitive at predicting ship seismic wave in shallow sea. Graphical abstract: Highlights: A hybrid superposition method is devised for accurately predicting ship seismic wave. The ship seismic wave is equivalently calculated by using superposition principle. In the developed method, original problem is reduced by one dimension. Truncating and discretizing the field space are not required for the prediction.