Abstract
Perforation operations are an important part of the well completion process in many field developments today. However, there are two considerations paramount to such a design. First, the post-job well production rate depends critically on a complex system in which the formation, perforations, and wellbore are dynamically and nonlinearly coupled together. Second, perforating operations can place a large dynamic load on downhole equipment. Predicting the spatial loading and its transient behavior is an important step in a completion design that implements risk mitigation to avoid damaging downhole equipment. In this study, we use a numerical modeling tool that has been used to simulate the dynamic behavior of the wellbore-reservoir system during perforating pre-job design. A typical case for this modeling software generates a quantitative prediction for the transient distribution of pressure, mass, velocity, and energy from the perforating event. These transients are commonly analyzed and incorporated into completion designs.
In this paper, two critical components of modeling relative to perforating job design will be evaluated and characterized. First, the wellbore-perforation-formation coupling will be evaluated and characterized using a modified version of the code designed to model API RP-19B Section IV flow testing. Section IV experimental data will be compared to numerical predictions. We present successful predictions of pressure transients that match experimental data. The second component of the software evaluated and characterized in this paper is the pressure, volume, temperature (PVT) accuracy and its implementation. NIST thermodynamic data, together with shock tube examples with pressure and temperature ranges relevant for deep-water completions will be used. We demonstrate that, although the current PVT implementation is accurate on typical downhole jobs, it could have limitations at higher pressures and temperatures. Finally, we present an improved numerical implementation for the modeling software.
