Search Results (3)

The fracture network generated by hydraulic fracturing in unconventional shale reservoirs contains numerous microfractures that are connected to macroscopic fractures. These microfractures serve as crucial pathways for shale gas to flow out from micro- and nano-scale pores, playing a critical role in enhancing shale gas recovery. Currently, more attention is being given by academia and industry to the evolution of macroscopic fracture networks, while the understanding of the microfracture mechanisms and evolution is relatively limited. A significant number of microfractures are generated during the hydraulic fracturing process of shale. These microfractures subsequently propagate, merge, and interconnect to form macroscopic fractures. Therefore, studying the fracture process of rock masses from a microscale perspective holds important theoretical significance and engineering value. Based on the authors’ research experience and literature review, this paper provides a brief overview of current progress in shale microfracture research from five aspects: in situ observation experiments of microfractures in shale, formation and evolution processes of discontinuous microfractures, the impact of inhomogeneity on microfracture propagation, measurement methods for microscale mechanical parameters and deformation quantities in shale, and numerical simulation of shale microfractures. This paper also summarizes the main challenges and future research prospects in shale microfracture studies, including: (1) quantitative characterization of in situ observation experimental data on shale microfractures; (2) formation and evolution laws of macroscopic, mesoscopic, and microscopic multi-scale discontinuous fractures; (3) more in-depth and microscale characterization of shale heterogeneity and its deformation and fracture mechanisms; (4) acquisition of shale micro-mechanical parameters; (5) refinement and accuracy improvement of the numerical simulation of microfractures in shale. Addressing these research questions will not only contribute to the further development of microfracture theory in rocks but also provide insights for hydraulic fracturing in shale gas extraction. Full article

Machine learning is the main technical means for lithofacies logging identification. As the main target of shale oil spatial distribution prediction, mud shale petrography is subjected to the constraints of stratigraphic inhomogeneity and logging information redundancy. Therefore, choosing the most applicable machine learning method for different geological characteristics and data situations is one of the key aspects of high-precision lithofacies identification. However, only a few studies have been conducted on the applicability of machine learning methods for mud shale petrography. This paper aims to identify lithofacies using commonly used machine learning methods. The study employs five supervised learning algorithms, namely Random Forest Algorithm (RF), BP Neural Network Algorithm (BPANN), Gradient Boosting Decision Tree Method (GBDT), Nearest Neighbor Method (KNN), and Vector Machine Method (SVM), as well as four unsupervised learning algorithms, namely K-means, DBSCAN, SOM, and MRGC. The results are evaluated using the confusion matrix, which provides the accuracy of each algorithm. The GBDT algorithm has better accuracy in supervised learning, while the K-means and DBSCAN algorithms have higher accuracy in unsupervised learning. Based on the comparison of different algorithms, it can be concluded that shale lithofacies identification poses challenges due to limited sample data and high overlapping degree of type distribution areas. Therefore, selecting the appropriate algorithm is crucial. Although supervised machine learning algorithms are generally accurate, they are limited by the data volume of lithofacies samples. Future research should focus on how to make the most of limited samples for supervised learning and combine unsupervised learning algorithms to explore lithofacies types of non-coring wells. Full article

Due to the advantages of good fracture performance and the application of carbon capture and storage (CCS), supercritical carbon dioxide (SC-CO₂) is considered a promising alternative for hydraulic fracturing. However, the fracture initiation mechanism and its propagation under pressurized SC-CO₂ jet are still unknown. To address these problems, a fluid–structure interaction (FSI)-based numerical simulation model along with a user-defined code was used to investigate the fracture initiation in an inhomogeneous shale rock. The mechanism of fracturing under the effect of SC-CO₂ jet was explored, and the effects of various influencing factors were analyzed and discussed. The results indicated that higher velocity jets of SC-CO₂ not only caused hydraulic-fracturing ring, but also resulted in the increase of stress in the shale rock. It was found that, with the increase of perforation pressure, more cracks initiated at the tip. In contrast, the length of cracks at the root decreased. The length-to-diameter ratio and the aperture ratio distinctly affected the pressurization of SC-CO₂ jet, and contributed to the non-linear distribution and various maximum values of the stress in shale rock. The results proved that Weibull probability distribution was appropriate for analysis of the fracture initiation. The studied parameters explain the distribution of weak elements, and they affect the stress field in shale rock. Full article