Seismic signal data processing method and device, electronic equipment and storage medium

Abstract

The invention discloses a seismic signal data processing method and device, electronic equipment and a readable storage medium. The method comprises the following steps: filtering original seismic signal data by using a singular spectrum analysis method to obtain initial filtering data; calculating the difference and similarity between the original seismic signal data and the initial filtering data; and calculating a weighting matrix based on the difference, the similarity, a preset weighting intensity control parameter used for predicting the energy attenuation of the unstable disturbance and a local similarity threshold used for determining the signal retention degree. And the original seismic signal data is filtered again based on the weighting matrix to obtain a final seismic signal processing result, so that the signal-to-noise ratio of the seismic signal is effectively improved, and the seismic noise can be suppressed more efficiently and stably.

Description

technical field [0001] The present application relates to the technical field of signal processing, and in particular, to a data processing method, apparatus, electronic device and readable storage medium for seismic signals. Background technique [0002] Noise in seismic data will bring errors to subsequent attribute analysis, wave impedance inversion, fracture prediction, etc., which will seriously affect the interpretation of subsurface structures and geophysical inversion. Since the seismic data collected in the field will inevitably be disturbed by noise, it is necessary to improve the ratio of signal to noise in the signal system, that is, the signal-to-noise ratio. [0003] In order to improve the signal-to-noise ratio of the seismic data, the incoherent noise in the acquired seismic signal data can be attenuated. The singular spectrum analysis method can detect the low-rank structure of the data. Based on the different performances of the seismic signal and random n...

AI Technical Summary

Problems solved by technology

However, because the fitting curve of the quadratic form of the singular spectrum analysis method is very sensitive to non-Gaussian interference, when the seismic data is disturbed by unstable noise, the performance of the singular spectrum analysis method is not stable, and the weighting criterion based on the absolute difference cannot identify and suppress outliers, it is very easy to cause a large amount of noise energy to remain or be regarded as interference. The denoising effect of seismic data is not obvious, and the signal-to-noise ratio is not high. However, if n (n>2) iterations are used to obtain better denoising effect, which in turn leads to less efficient data processing due to the greater computational cost required for iterative steps, especially when robust functions are difficult to approximate

Images