摘要
By integrating laboratory physical modeling experiments with machine learning-based analysis of dominant factors,this study explored the feasibility of pulse hydraulic fracturing(PHF)in deep coal rocks and revealed the fracture propagation patterns and the mechanisms of pulsating loading in the process.The results show that PHF induces fatigue damage in coal matrix,significantly reducing breakdown pressure and increasing fracture network volume.Lower vertical stress differential coefficient(less than 0.31),lower peak pressure ratio(less than 0.9),higher horizontal stress differential coefficient(greater than 0.13),higher pulse amplitude ratio(greater than or equal to 0.5)and higher pulse frequency(greater than or equal to 3 Hz)effectively decrease the breakdown pressure.Conversely,higher vertical stress differential coefficient(greater than or equal to 0.31),higher pulse amplitude ratio(greater than or equal to 0.5),lower horizontal stress differential coefficient(less than or equal to 0.13),lower peak pressure ratio(less than 0.9),and lower pulse frequency(less than 3 Hz)promote the formation of a complex fracture network.Vertical stress and peak pressure are the most critical geological and engineering parameters affecting the stimulation effectiveness of PHF.The dominant mechanism varies with coal rank due to differences in geomechanical characteristics and natural fracture development.Low-rank coal primarily exhibits matrix strength degradation.High-rank coal mainly involves the activation of natural fractures and bedding planes.Medium-rank coal shows a coexistence of matrix strength degradation and micro-fracture connectivity.The PHF forms complex fracture networks through the dual mechanism of matrix strength degradation and fracture network connectivity enhancement.
基金
Supported by the National Natural Science Foundation of China(52274014,52421002).
