Dimensionality reduction adaptive interlayer multiple suppression method for land seismic exploration data

Abstract

The invention belongs to the exploration seismic signal processing field and relates to an adaptive prediction and suppression method of land seismic exploration data inter-layer multiple waves, in particular, a dimensionality-reduction adaptive inter-layer multiple wave suppression method of land seismic exploration data. According to the method, inter-layer multiple wave prediction and suppression of the land seismic exploration data are realized in a pre-stack CMD (Common Middle Point) trace gather through using a virtual event method. The method includes the steps of input data preparation, reference trace construction and multiple wave suppression. According to the method, when dynamic correction is adopted to eliminate travel related to an offset distance caused by the distribution of shot check points, the method has no requirements for the distribution of the shot check points; two-dimensional or three-dimensional data are converted into one-dimensional data through dimensionality reduction, so that calculation efficiency is improved; and weighted stacking is performed, so that reference traces can be obtained, and therefore, the noise immunity of a system can be improved.

Description

technical field [0001] The invention belongs to the field of seismic signal processing for exploration, and relates to an adaptive prediction and suppression method for interlayer multiple waves of land seismic data. A virtual event method is applied to seismic pre-stack common medoid (CMP) gather data to realize land seismic data Prediction and suppression of interlayer multiples. Background technique [0002] Multiples are common correlated noise in prestack seismic data and are developed in marine seismic exploration and many deep land explorations. Commonly used seismic data imaging technologies are based on the assumption of primary reflected wave signals. Since multiple waves do not meet the imaging conditions, they are regarded as coherent noise, which often affects the amplitude, phase and frequency of the primary wave generated by the exploration target layer, and Structural artifacts are generated to cover up the real structure of deep formations, reducing the rel...

AI Technical Summary

Problems solved by technology

However, it is difficult to realize the prediction of interlayer multiples in pre-stack data mainly due to the following reasons: (1) For land seismic data, the signal-to-noise ratio of pre-stack gathers is extremely low, which affects the accuracy of multiple ratio prediction; (2) Predictive subtraction methods, such as SRME and virtual event method, require the use of regularized data and a sufficiently dense distribution of offset points, but observation systems with complex surface conditions are often difficult to meet these requirements

Images