Accurately establishing a 3D digital core model is of great significance in oil and gas production. The physical experiment method and numerical modeling method are common modeling methods. With the development of deep learning technology, a variety of deep learning algorithms have been applied to digital core modeling. The digital core modeling method based on generative adversarial neural networks (GANs) has attracted wide attention due to its good quality and simple generation process. The disadvantage of this method is that the network needs thousands of trainings to achieve acceptable results. For this reason, this paper proposes to use the pretrained GANs for digital core modeling training, which can greatly reduce the number of network training while ensuring the core modeling effect. We can use the presented method to quickly complete the training and use the trained generator model to obtain multiple digital cores. By analyzing the quality of the generated cores from multiple aspects, it is revealed that the properties of the generated cores are in good agreement with the ones of the real core samples. The results indicate the reliability of the pretrained GAN method.

Marine shale gas exploration targets are reaching into shallow to deep or ultradeep burial depths. Rock microstructure and reservoir quality emerge as the main risk considerations for a profitable reservoir at these large depths. Shallow to deep marine shales occur in the Silurian Longmaxi Formation of the Weiyuan Block, which is located on the southwestern margin of the Sichuan Basin. However, few details of the characteristics of the Longmaxi shales in this Block have been reported. In this study, five wells approximately 3500 m deep were drilled. Field emission scanning electron microscopy, low-pressure gas adsorption and core plug porosity-permeability measurements are conducted on 6 shallow (2651–2940 m) and 11 deep (3539–3575 m) Longmaxi samples to obtain the organic geochemical characteristics, mineral constitutions, pore structures and petrophysical properties, and they are the major controls on reservoir quality. The results show similar mineralogical and organic geochemical characteristics in all samples from the various depths. Both shallow and deep shales are mainly composed of quartz, carbonate and clays and have a total organic carbon (TOC) content more than 2 wt.% and a mean S1 + S2 value more than 52 mg/g. Source rock quality criteria using the TOC and S1 + S2 suggest most shale samples fall excellent source rocks. The samples are mostly siliceous rocks that contain organic pores, intraparticle dissolution pores, interparticle quartz pores and interparticle clay pores and they play a positive role in improving reservoir quality. Pore surface area and pore volume increase with increasing TOC, indicating that the porous organic fraction is a major control on pore structure and porosity. We suggested that siliceous deep Longmaxi formation in Weiyuan Block belongs to a high-quality shale with an average TOC value of 3.9 wt.% and well-developed pore networks, which should be the important target for deep shale gas exploration.