Seismic horizon identification and tracking method and system

Abstract

The invention provides a seismic horizon identification and tracking method and system. The method comprises logging data and actual seismic data are processed to obtain a training data set and tag data; a pre-established multi-scale time series deep learning model is trained through utilizing the training data set and the tag data; the trained multi-scale deep learning model is utilized for automatic horizon tracking, seismic data interpretation is performed through utilizing the multi-scale time series deep learning model based on the expanded convolutional neural network and the time-seriesrecurrent neural network, the method adapts to the position where the seismic horizon property changes rapidly, and accuracy of the result is greatly improved.

Description

technical field [0001] The invention belongs to the technical field of seismic exploration, and in particular relates to a method and system for identifying and tracking seismic horizons, computer equipment, and a computer-readable medium. Background technique [0002] The sequence of appearance of the strata is related to the sedimentary history of the crust. The sediments of different geological ages have different characteristics, and the corresponding seismic data features and sequence from shallow to deep are also regular. The main work of seismic interpretation is to process, analyze and study seismic data, and then infer the properties of underground structures and rock formations, and look for possible oil-bearing structures, such as faults. When the underground rock layer is in operation in the crust, it will be broken and dislocated when the force reaches a certain intensity, thus forming a fault. . Seismic horizon interpretation is the basic work of seismic inte...

AI Technical Summary

Problems solved by technology

[0005] In view of this, the present invention provides a seismic horizon identification and tracking method, system, computer equipment, and computer-readable medium to solve the problem of linear algorithms in the prior art that do not adapt to locations where seismic horizon attributes change rapidly and ignore seismic data. Longitudinal time series information leads to low accuracy of seismic horizon tracking and identification

Images