Wave impedance inversion method using neural network and neural network system

Abstract

The invention relates to a wave impedance inversion method using a neural network and a neural network system. The neural network takes N channels of seismic data composed of i-th channel seismic data and multiple channels of seismic data adjacent to the i-th channel seismic data and i-th channel initial model data as input, determines i-th channel wave impedance data as output, and comprises: N parallel feature extraction layers, wherein each feature extraction layer is configured to extract time sequence features of a channel of seismic data input to the feature extraction layer; a merging layer configured to adaptively merge the time sequence features output by the N feature extraction layers to obtain spatial-temporal features of the N channels of seismic data; a regression layer configured to map the spatial-temporal features from a feature domain to a target domain; and an output layer configured to determine the i-th channel wave impedance data according to the output of the regression layer and the i-th channel initial model data. According to the invention, the accuracy of wave impedance inversion is higher, the continuity is stronger, the anti-noise performance is better, and a good inversion effect can still be kept when the initial model is not accurate.

Description

technical field [0001] The present application relates to the technical field of oil and gas exploration, in particular to a wave impedance inversion method using a neural network and a neural network system. Background technique [0002] Seismic wave impedance inversion is a method of recovering broadband wave impedance data from seismic data with limited frequency bandwidth by comprehensively utilizing existing geological and logging data under the guidance of seismic data. It has been widely used in oil and gas exploration. Reservoir qualitative and quantitative prediction, well pattern deployment, reserve calculation, reservoir dynamic monitoring, etc. in the oil and gas development stage. Because the actual geophysical problems are very complex, and our understanding of them is usually very vague, which makes most of the models established when inverting wave impedance are approximate. The neural network can learn the hidden knowledge from a large amount of existing da...

AI Technical Summary

Problems solved by technology

When the underground structure is complex, the relationship between seismic data and inversion parameters is very complicated, and the network generally cannot express this relationship accurately. Although the learning ability of the network can be improved by increasing the depth of the network, due to the limitation of the amount of data, this will increase the risk of network overfitting

Images