Method, device, equipment and system for predicting gas bearing capacity of tight sandstone reservoir

Abstract

The embodiment of the invention discloses a method, device, equipment and system for predicting the gas bearing capacity of a tight sandstone reservoir. The method comprises the following steps: acquiring pre-stack seismic angle gather data of a tight sandstone reservoir to be predicted; inputting the pre-stack seismic angle gather data into a pre-constructed multi-attribute simultaneous prediction hybrid convolutional neural network, wherein the multi-attribute simultaneous prediction hybrid convolutional neural network comprises a shared sub-network and a branch network of a plurality of tasks; based on the shared sub-network, obtaining multi-source fusion information of the tight sandstone reservoir to be predicted; and respectively inputting the multi-source fusion information into branch networks of the plurality of tasks, and simultaneously obtaining prediction results of tight sandstone lithofacies distribution and reservoir gas bearing capacity. By utilizing the embodiment of the invention, the precision of predicting the gas bearing capacity of the tight sandstone reservoir from the pre-stack seismic angle gather data based on the convolutional neural network can be effectively improved, the network complexity and the storage space are reduced, and the calculation efficiency is improved.

Description

technical field [0001] The embodiment plan of this specification belongs to the new field of machine learning and the field of exploration geophysical processing and interpretation in geosciences, and especially relates to a method, device, equipment and system for predicting gas-bearing properties of tight sandstone reservoirs. Background technique [0002] my country has abundant tight sandstone gas resources and huge exploration potential. It is one of the main exploration and development goals for my country's petroleum industry to maintain stable or increase production at this stage, but the proven rate is relatively low. Compared with tight sandstone gas in the United States and conventional sandstone gas in my country, tight sandstone gas reservoirs in China have the characteristics of low porosity, low permeability, complex pore structure, poor distribution stability and strong heterogeneity. Due to the complexity of tight sandstone gas reservoirs, the seismic respon...

AI Technical Summary

Problems solved by technology

However, these practical technologies generally require conditions such as "the actual observation data satisfy the principle of plane wave superposition, the actual underground medium is completely elastically isotropic, and the difference in the elastic parameters of adjacent formations is small", which limits the application range of gas-bearing prediction.

Moreover, from the perspective of information volume, linearization loses the large-offset nonlinear seismic information components that are more sensitive to gas-bearing properties.

Furthermore, the constraints of log, geological or electromagnetic field information that are homogeneous to seismic but heterogeneous are rarely considered in the inversion process

Images