Seismic data surface wave suppression method based on generative adversarial network

Abstract

The invention provides a seismic data surface wave suppression method based on a generative adversarial network. The method mainly comprises: creating a group of training data obtained through data processing; providing the training data for a neural network, training a generator and a discriminator, enabling the data generated by the generator to cheat the discriminator through network training,and enabling the discriminator to identify the data generated by the generator and the training data; and during a network training process, using a generator and discriminator mutual confrontation training method to enable the generator to obtain good data and to enable the discriminator to perform more accurate identification; and after the network training is completed, providing original seismic data containing surface waves for the generator, and generating, by the generator, seismic data without surface wave. Analog data and actual seismic data processing examples show that the method provided by the invention can be used for effectively suppressing the surface waves in the seismic data, protecting effective signals, and improving the signal-to-noise ratio of the seismic data.

Description

technical field [0001] The invention relates to a seismic data processing method, in particular to a method for suppressing surface waves in seismic data based on a generated countermeasure network. Background technique [0002] In the process of field seismic data acquisition, due to the influence of the construction environment and underground seismic geological conditions, there are many types of noise in the original seismic data collected in the field, such as: surface waves, multiple waves, random noise, mechanical interference, alternating current, etc. Interference waves such as interference and a small amount of abnormal amplitude. These interference waves can generally be divided into two categories, coherent noise and random noise. Random noise refers to the random occurrence in seismic data, which has no fixed frequency, nor fixed propagation direction and speed. Coherent noise means that it has a certain regularity in time, has corresponding kinematic characte...

AI Technical Summary

Problems solved by technology

[0005] In summary, the prior art methods are only suitable for suppressing the original simple underground structure, and are not suitable for suppressing surface wave noise in seismic data of complex underground structures

Images