Data driving method and system for seismic wave first arrival pickup

Abstract

The invention discloses a data driving method and system for seismic wave first arrival pickup, and relates to the technical field of exploration geophysics, and the method comprises the steps: carrying out the linear superposition summation of all first virtual seismic traces formed at the position of a stationary-phase seismic source, and obtaining a first linear superposition result; performing phase weighted stacking on the first linear stacking result by using a first weighting coefficient to obtain a second virtual seismic trace; obtaining a super virtual seismic trace according to the second virtual seismic trace; performing linear superposition summation on all the super virtual seismic traces generated at the stable-phase detection point position to obtain a second linear superposition result; and performing phase weighted stacking on the second linear stacking result by using a second weighting coefficient to obtain a final super virtual seismic trace of the target trace gather in the target shot gather. According to the invention, under the condition that the signal-to-noise ratio of the original data is very poor and the original data volume is limited, the signal-to-noise ratio of the seismic data first-arrival wave is obviously improved, so that high-quality first-arrival wave input data is provided for seismic wave first-arrival pickup.

Description

technical field [0001] The invention relates to the technical field of exploration geophysics, in particular to a data-driven method and system for picking up seismic wave first arrivals. Background technique [0002] Accurate picking of the first arrival wave of seismic data is very important for the processing of seismic data, and can provide reliable parameters for refraction wave static correction, tomography, interpretation of VSP data, and processing of cross-well seismic data. The accuracy of first-arrival picking directly affects the accuracy of subsequent data processing. [0003] There are many commonly used seismic wave first-arrival picking methods, such as correlation method, energy ratio method, fractal dimension method, neural network method, and image processing-based methods, etc. Picking the first arrivals of seismic data using the existing first-arrival picking methods has its own advantages and disadvantages in practical applications, but the common prob...

AI Technical Summary

Problems solved by technology

Picking the first arrivals of seismic data using the existing first-arrival picking methods has its own advantages and disadvantages in practical applications, but the common problem is poor noise immunity, which is difficult for seismic data with low signal-to-noise ratio. Accurately pick up the first arrival

Therefore, in order to ensure the accuracy and efficiency of first-arrival picking, a method that can effectively improve the signal-to-noise ratio and quality of first-arrival seismic data is needed to solve the problem that conventional first-arrival picking methods are difficult to accurately pick first-arrivals for low-SNR seismic data. question

[0005] At present, there is a data-driven method for first-arrival picking of seismic waves. This method depends on the number of gathers involved in the calculation. The signal-to-noise ratio of first-arrival seismic data can be improved through correlation, convolution, and superposition, so that the first-arrival Picking provides high-quality first-arrival input data, but when the quality of the original data is poor, that is, the signal-to-noise ratio of the original data is poor, and the amount of original data is limited, the ability of this method to improve the signal-to-noise ratio is limited, and the effect of improving the signal-to-noise ratio is not ideal

Images