Adaptive Subtraction Based on Expanded Multichannel U-Net With Multipattern Multiple Model for Surface-Related Multiple Removal

Adaptively subtracting multiple models from the initial data is an essential assignment for the successful elimination of seismic surface-related multiples. The conventional expanded multichannel linear regression (EMLR) method has been proposed to address this challenge by utilizing a multipattern multiple model. These patterns include the multiple model itself and its first derivative, its Hilbert transform, and its first derivative of the Hilbert transform, which are matched with the initial data in the EMLR method. It may lead to inaccurate primary preservation or give rise to residual multiples by using the linear regression (LR) model. The existing U-Net method effectively mitigates complex disparities between the multiple model and actual multiples by integrating adaptive subtraction into the non-LR architecture. Nevertheless, residual multiples are produced by this method using the multiple model itself, especially in complex media contexts. In order to improve surface-related multiple removal accuracy, we propose the expanded multichannel U-Net (EMUN) method with a multipattern multiple model. In the proposed method, the initial data are matched with the multipattern multiple models through U-Net in the way of self-supervised training without true primaries as labels. The proposed method incorporates rich information from the expanded multichannel of U-Net, enabling better U-Net training for adaptive subtraction. In contrast to the EMLR method and the existing U-Net method, the proposed EMUN method exhibits exceptional efficacy in protecting primaries and eliminating surface-related multiples, as evidenced by its outstanding performance in both synthetic and field data tests.