Optimizing Well Trajectories for Enhanced Oil Production in Naturally Fractured Reservoirs: Integrating Particle Swarm Optimization with an Innovative Semi-Analytical Model Framework

Summary Well trajectory optimization is a crucial component in the drilling engineering of naturally fractured reservoirs. The complex heterogeneity and anisotropy of such reservoirs significantly affect the pressure drop distribution within the well and, consequently, the oil well’s output, impacting the economic benefits of the well. Therefore, optimizing the well segment trajectory is key to efficient reservoir development. However, traditional well trajectory optimization methods primarily focus on geological structures and drilling engineering costs, often overlooking future production benefits of the oil well. This paper proposes a new method that first establishes a semi-analytical production prediction model capable of describing complex well trajectories. Although the semi-analytical model has unique advantages in well trajectory description, it typically treats the reservoir as a homogeneous entity, which complicates handling complex reservoir characteristics. To overcome this limitation, we combined optimization algorithms and neural networks to construct a framework for addressing reservoir heterogeneity (Semianalytical Model Framework for Unconventional Wells in Heterogeneous Reservoirs, USAMF-HR), enhancing the semi-analytical model’s ability to describe reservoir heterogeneity. Building on this framework, we applied the particle swarm optimization (PSO) algorithm and introduced constraints on the rationalization of initial well trajectories, as well as limits on particle movement speed and displacement, with the maximization of net present value (NPV) as the objective function, to optimize well trajectory coordinates. Through specific case analysis, the reasonableness and practicality of this method have been verified. The results show that this method can quickly and effectively plan the optimal well trajectory, significantly increasing productivity while reducing costs.