Prediction of Temperature Distribution in Intelligent Wells

Abstract

Abstract Intelligent well system technology enables downhole monitoring and zonal, fluid production control in real time. This allows well control decisions to be implemented that optimize recovery and avoid future problems. Currently, both pressure and temperature are usually monitored downhole; while often only the pressure data is used to provide the key information about the downhole performance. Temperature measurements have the potential to be as informative as pressure measurements in reflecting the reservoir's and the well's production performance. However, currently available models are unable to simulate the temperature profile of an intelligent well correctly. This paper provides a theoretical underpinning for the temperature data interpretation workflow. A novel temperature model for multiphase flow with (possible) phase changes in an intelligent completion will be presented. Reservoir and sandface temperature performance is also analyzed and coupled with the intelligent well temperature model. Three example cases illustrate the temperature model's utility. They demonstrate that a small, but significant temperature response can be observed in the complex system of an intelligent well equilibrated with a reservoir zone. The magnitude of the signal used for influx detection is quantified for wells with different inclinations and influx locations. The importance of measurement resolution is also discussed. This is a building block for a method to detect gas and/or water influxes and obtain phase flow rates at various locations within an intelligent well, is an important step towards a comprehensive well management scheme. 1.0 Introduction Intelligent wells, equipped with downhole interval flow control devices, allow production optimization for reservoirs of different complexity and flow conditions. Downhole monitoring systems in such wells give the possibility to trace their inflow performance in real time and to take appropriate, early control decisions to avoid future, technical problems. Downhole data can also be used to update the reservoir model, leading to a better description of the reservoir, an improved performance prediction and to a more effective, reservoir depletion strategy. An intelligent well is thus a powerful instrument in the reservoir management tool box. Development of measurement devices such as permanently installed, downhole, point measurement gauges or distributed temperature sensors has lead to the ability to make pressure and temperature measurements with a high temporal and spatial resolution. Location of such gauges or sensors along an inflow interval potentially gives the opportunity to track the inflow performance distribution in real time 1. However, this requires the availability of an interpretation method for the measured data that can calculate the pressure and temperature profiles along an intelligent completion. Further, analysis and understanding of how these profiles depend on the inflow can be used to interpret them both qualitatively (e.g. for breakthrough detection) as well as quantitatively (e.g. for rate allocation). Pressure calculation and analysis is, in some circumstances, a relatively straightforward problem since currently available pressure equations and correlations can be reliably applied to the area of interest 2, 3. However, temperature calculations, the subject of this paper, need further attention. This paper presents a calculation method for the temperature distribution along the wells with multizone intelligent (or simpler) completion under conditions of multiphase flow in which the reservoir-well thermal interaction is fully accounted for. It incorporates the necessary equations to calculate the temperature profile of the full reservoir-well system, i.e. in the reservoir, along the completion and across interval control valves (Figure 1). The derivation of these equations is given. The method will be used to describe the behavior of three intelligent well cases. It is a building block in the construction of an accurate, downhole data interpretation and soft-sensing tool for influx detection, tracking and quantification.