A 3D attention U-Net network and its application in geological model parameterization

To solve the problems of convolutional neural network-principal component analysis (CNN-PCA) in fine description and generalization of complex reservoir geological features, a 3D attention U-Net network was proposed not using a trained C3D video motion analysis model to extract the style of a 3D model, and applied to complement the details of geologic model lost in the dimension reduction of PCA method in this study. The 3D attention U-Net network was applied to a complex river channel sandstone reservoir to test its effects. The results show that compared with CNN-PCA method, the 3D attention U-Net network could better complement the details of geological model lost in the PCA dimension reduction, better reflect the fluid flow features in the original geologic model, and improve history matching results.

Automated summary

To address the issues with CNN-PCA in accurately describing and generalizing complex reservoir geological features, a 3D attention U-Net network was proposed in this study. This network does not rely on a pre-trained C3D video motion analysis model, but instead focuses on complementing the details lost in the dimension reduction process of PCA. The experimental results on a complex river channel sandstone reservoir demonstrate that the 3D attention U-Net network outperforms the CNN-PCA method in capturing the missing details and reflecting fluid flow features, leading to improved history matching results.