Seismic waveform clustering method and device

Abstract

The invention provides a seismic waveform clustering method and device, and the method comprises the steps of: determining seismic waveform data and a seismic facies label of the seismic waveform data according to drilling and logging information; delineating a set area in a target work area, and extracting label-free seismic facies waveform data of the set area; determining labeled seismic facies data according to the seismic waveform data and the unlabeled seismic facies waveform data; performing multi-scale discretization processing on the labeled seismic facies data, and determining multi-scale labeled waveform data; training a recurrent neural network model according to the multi-scale labeled waveform data and the seismic facies label of the seismic waveform data, and determining the trained recurrent neural network model; and inputting the seismic trace of a target work area into the trained recurrent neural network model, and determining a seismic facies classification result of the target work area. According to the invention, the semi-supervised learning method of the recurrent neural network model generative adversarial network is used to realize rapid and precise classification of the seismic facies of the target work area.

Description

technical field [0001] The invention relates to the technical field of oil and natural gas development, in particular to a seismic waveform clustering method and device. Background technique [0002] The overall change of seismic waveform is a comprehensive reflection of seismic wave amplitude, frequency, and phase, and is an important seismic attribute parameter. In the process of exploration and development of complex lithology reservoirs, seismic waveform classification technology is an effective and fast method to predict the spatial distribution of reservoirs. [0003] Seismic waveform classification in the prior art mainly uses neural network technology or cluster analysis technology to classify the shape of the seismic trace (the overall change of the seismic signal). First, several typical seismic trace shapes are divided, and each actual seismic trace is endowed with a typical shape based on similarity; the neural network trains the actual seismic traces in a given...

AI Technical Summary

Problems solved by technology

However, due to the sparseness of well logging data relative to seismic data, well logging data can only represent local geological information. In traditional supervised classification methods, the classification results are often poor.

Images