Progress in brittleness evaluation and prediction methods in unconventional reservoirs
CAO Dongsheng, ZENG Lianbo, LYU Wenya, XU Xiang, TIAN He
1 State Key Laboratory of Petroleum Resource and Prospecting in China University of Petroleum-Beijing, Beijing 102249, China 2 College of Geoscience, China University of Petroleum-Beijing, Beijing 102249, China
Brittleness is of great significance for deep rock engineering and resource development, especially for unconventional oil and gas resources. Brittleness evaluation, fundamental principles of its prediction and research progress are summarized and analyzed. The structural characteristics of rock such as lithologic composition, bedding and damage, pore fluid and its occurrence characteristics, confining pressure, temperature, rock mass measurement scale, and stress path all impact brittleness. High brittleness unconventional reservoirs are characterized by a high content of brittle minerals, high Young's modulus, small total strain before fracture, dissipated energy in the pre-peak stage of the stress-strain curve, little fracture energy in the post-peak stage, low ductility, large internal friction angle and easy formation of complex fracture network systems in hydraulic fracturing. Unconventional reservoir brittleness research should focus on the formation frangibility and the ability to form complex fracture network systems. According to the types of data, brittleness evaluation methods mainly include mechanical experiment evaluation and evaluation based on logging and drilling data. Brittleness prediction is mainly based on prestack seismic inversion. The study of brittleness anisotropy and controlling factors help optimize and improve evaluation and prediction methods for different types of unconventional reservoirs. Due to the different research ideas and data sources, the applicability of different methods is also different. The integration and mutual verification of multiple data and methods is an important future development direction. Artificial intelligence, including machine learning algorithms, can organically integrate multiple data, collate effective information, and has the advantage of being more efficient and accurate. Artificial intelligence is promising in geological research controlled by multiple nonlinear factors such as reservoir brittleness.
Key words：
unconventional reservoir; factors affecting brittleness; brittleness evaluation method; brittleness prediction method; progress in brittleness research
