Objective The multistage horizontal well fracturing based on multi-cluster perforation has emerged as an effective approach for the efficient production of deep coal-measure gas (CMG). However, deep coals feature unique rock mechanical properties, fluid loss characteristics of reservoirs, and microfracture developmental characteristics, which differ significantly from those of unconventional shales, tight sandstones, and shallow coals. As a result, existing fracturing experience cannot be directly applied, highlighting the investigation of fracturing parameter optimization tailored for deep CMG production.

Methods Focusing on deep CMG reservoirs in the Baijiahai area within the Junggar Basin, this study constructed a model of multistage horizontal well fracturing based on multi-cluster perforation for a composite geological structure composed of a roof, a floor, gangue, and coals. This model considered the effects of bedding and cleats in coals on fracture propagation and the fluid loss of reservoirs. Furthermore, critical reservoir parameters in the model were corrected using data derived from mini-fracture tests. These contributed to an elevated accuracy of the model, which, therefore, allowed for the comprehensive characterization of the physical and mechanical properties of deep CMG reservoirs. The optimal fracturing fluid system for deep CMG reservoirs was selected through numerical simulation, and a method for designing the optimal pumping parameter combination under multiple objectives was developed. Finally, the simulation results were verified using microseismic data.

Results and Conclusions The results indicate that compared to the low-viscosity slickwater and gelled systems, the variable-viscosity slickwater system was more suitable for the fracturing of deep CMG reservoirs while also enjoying advantages in terms of fracture length, fracture width, and fracture equilibrium. The optimal parameter combination for fracturing in the target block was determined to include three clusters per stage, a cluster spacing of 19 m, a proppant volume of 2.8 m³ per meter, and an injection rate of fracturing fluids of 16 m³/min. The microseismic monitoring results of test wells demonstrate that after parameter optimization, the fracture half-length and stimulated reservoir volume (SRV) increased by 57.2 % and 12.3 %, respectively. A comparison of fracture geometries under varying fracturing parameter combinations reveals that in deep CMG reservoirs subjected to severe blocking by barriers, fractures typically cannot propagate in their height directions, necessitating increasing their lengths and widths. Increasing the proppant volume and the injection rate of fracturing fluids can significantly increase fracture lengths and widths, playing a key role in improving reservoir fracturing performance.