Wave field separation principle based denoising method for increasing signal to noise ratio of seismic data

Abstract

The invention provides a denoising method for increasing the signal to noise ratio of seismic data, and belongs to the field of geophysical prospecting for petroleum. The data partitioning technique and the wave field separation technique are combined to suppress stochastic noise so as to increase the signal to noise ratio of seismic data. The method includes: partitioning the seismic data, subjecting each data block to complex field SVD (singular value decomposition) filtering, using the wave field separation technique for decomposition of characteristic values, using optimized characteristic values for data reconstruction, and replacing original frequency slices to realize suppression of the stochastic noise. Compared with the conventional SVD technique which assumes horizontal lineups, the method has the advantages that any relative time difference correction for the lineups is not needed, and denoising effects are stable. By means of the partitioning technique, unsteady state of inclination angle variation in actual seismic data can be overcome effectively, and time occupied by SVD and reconstruction is greatly shortened.

Description

technical field [0001] The invention belongs to the field of petroleum geophysical exploration, and in particular relates to a denoising method for improving the signal-to-noise ratio of seismic data based on the principle of wave field separation. Background technique [0002] Random noise is a main type of interference wave encountered in seismic data processing. In seismic records, it mainly manifests as disordered waveform, wide frequency band, uncertain apparent velocity, and no definite propagation direction. It can satisfy statistical laws to a certain extent. The existence of random noise will reduce the signal-to-noise ratio of seismic data and affect the quality of data processing. The methods of suppressing random noise are mainly divided into 3 categories: [0003] 1. Methods based on prediction techniques (f-x deconvolution and projection deconvolution, etc.); [0004] 2. Methods based on statistical principles (median filtering and polynomial fitting, etc.); ...

AI Technical Summary

Problems solved by technology

These techniques are computationally intensive and require complex corrections to apply to dip data

Moreover, the noise suppression technology based on SVD is implemented in the t-x domain, and generally requires the signal to be horizontal, which is not suitable for the situation where the events are interleaved.

Images