CO2 Pre-Fracturing Full Life-Cycle Simulation and Parameter Optimization Considering Geomechanics and Dynamic Fractures in Shale Oil Reservoirs

Abstract

Hydraulic fracturing is essential for shale oil development, but conventional water-based fracturing fluids may cause formation damage and environmental concerns because of incomplete flowback. Employing CO2 as a prefracturing fluid is a feasible and promising approach. However, the enhanced oil recovery (EOR) performance of CO2 prefracturing is influenced by multiple interacting geological and operational factors. In particular, parameter optimization requires consideration of both complex phase-behavior variations and the effects of in situ stress on fracture propagation. In this study, a multiphase, multicomponent numerical simulation model considering geomechanics and dynamic fracture was established. The model is used to quantify the effects of key geological and engineering parameters on production performance, and different machine-learning algorithms combined with random search are applied to determine optimal operating conditions. Results show that, below the minimum miscibility pressure (MMP), the EOR effect of CO2 prefracturing is mainly caused by oil swelling and interfacial-tension reduction, whereas above the MMP, miscibility and extraction dominate. Compared with water-based fracturing, CO2 prefracturing yields longer fractures, stronger reservoir-pressure support, better flowback performance, and higher oil recovery. Among the evaluated factors, bottomhole flowing pressure (BHFP), reservoir thickness, and CO2 injection rate have the greatest impact on performance. The optimal BHFP, CO2 injection rate, and fracture conductivity are 3.3 MPa, 7.7 m3/min, and 46 mD·m, respectively, resulting in up to 1.9 times higher production. The proposed workflow provides a practical method for evaluating and optimizing CO2 prefracturing in shale oil reservoirs. The results offer theoretical guidance for well selection and field-scale treatment design.