Fault identification plays a vital role in geologic structure interpretation, reservoir characterization, and reservoir evaluation. The traditional process for fault identification is label intensive. Deep learning-based automatic fault detection from seismic data has attracted extensive attention in recent years. However, using a 2D neural network (i.e., a single seismic data slice) leads to the loss of correlation among seismic slices, and using a 3D neural network (i.e., a 3D seismic data cube) increases the workload of fault labeling and the difficulty of training. To overcome the drawbacks of 2D and 3D neural networks, we have introduced a channel attention mechanism in the original U-net architecture. The new model, which is referred to as the 2.5D channel attention U-net (2.5D CAU-net), takes four adjacent seismic slices as inputs instead of the entire 3D seismic data cube. The efficacy of 2.5D CAU-net for fault detection is demonstrated using publicly available synthetic seismic data sets and real seismic field data sets. We explored the influence of cropping the original seismic data into various sizes on fault identification. Experimental results find that the 2.5D CAU-net can effectively use the correlation information among adjacent seismic slices to improve fault detection performance. For different seismic data cropping strategies, results indicate that the more extensive size cropping method allows the network to obtain more contextual information and more complete fault features, improving the efficacy of fault picking. However, the cropping approach with a larger size leads to a smaller number of labeled examples. When the amount of training data is not large enough, the larger size cropping approach tends to lead to model overfitting, which will decrease the performance of the fault recognition, especially in a 3D neural network. Overall, 2.5D CAU-net can improve the efficiency of fault recognition with less annotation effort and computational resources.

Low-to-medium maturity shale oil resources hold significant potential, but their economic accessibility is limited by low porosity, low permeability, and a low proportion of movable oil. In-situ conversion technology can crack organic matter and in-place oil into lighter molecules, enhancing oil and gas mobility and improving recovery rates. The success of this approach depends on dynamically evaluating the amount of movable shale oil during in-situ conversion. This study targets the lower submember of the fourth member of the Eocene Shahejie Formation (Lower Sha4 Member) in the Damintun Sag, Bohai Bay Basin, China. Through thermal simulation experiments, organic geochemical experiments, and nuclear magnetic resonance experiments, shale residual oil evaluation, organic matter hydrocarbon generation process evaluation, and dynamic evaluation of immovable oil were carried out. By integrating numerical simulations of the temperature field with the experimental results, a dynamic evaluation method for movable resources in shale during in-situ conversion was established. The findings indicate that the conversion rates of kerogen-to-oil and kerogen-to-gas first increase and then gradually stabilize as thermal maturity increases, with oil generation reaching its peak when the vitrinite reflectance reaches 1.0%. Long-term preserved shale samples were identified to contain immovable oil, the content of which increases with maturity before peaking and then declining. In-situ conversion of low-to-medium maturity shale in the upper part of the model can significantly increase movable oil resources in a year, potentially reaching the levels of extractable medium-to-high maturity shale. This work presented a crucial approach for assessing and improving in-situ conversion technology, providing a means of maximizing economic feasibility. Document Type: Original article Cited as: Bai, G., Chen, G., Cai, Z., Yan, H., Lu, S., Radwn, A. E. Movable oil content evaluation in low-to-medium maturity lacustrine shale during in-situ conversion. Advances in Geo-Energy Research, 2025, 16(1): 77-90. https://doi.org/10.46690/ager.2025.04.08