Field Testing of Analytical Techniques for Flow Estimation and Flow Allocation

Abstract

Abstract Fields with multiple producing reservoir units offer some of the most interesting asset development challenges. Several of these fields are now being completed with intelligent wells that enable commingling of multiple units with reservoir-control functionality. Unit production estimation and allocation of downhole flow rates are critical components of efficient reservoir and asset management. Deploying zonally dedicated subsurface flow meters is the most common way of estimating unit production rates on a real-time basis from an intelligent well. This paper presents results from analytical techniques for estimating unit production rates in real time by combining well architecture information, downhole pressure, and temperature data with appropriate reservoir inflow and interval control valve (ICV) flow equations. Since downhole pressure, temperature, and ICV information is already available in an intelligent well, this technique provides a lower-cost option for obtaining zonal and total well production and injection rates. The methodology used incorporates analytical choke equations, tubing performances, and nodal analysis (inflow performance relationships) with other reservoir parameters to build a flow estimation algorithm and model. Various downhole equipment (interval control valves, packers, pressure and temperature sensors, etc.) as well as related well information are brought into the system to set initial and final boundary conditions. Well-test data can be used to calibrate the system and improve the accuracy of the model. Field data from several wells have been run through the model; well tests from the field were used to calibrate and improve accuracy. Results vary from well to well. The system delivers flow-rate estimates greater than 90-percent accuracy when compared to actual flow-rate measurements from well tests and flow meters for some of the wells. The results show an operating envelope that covers a range of pressure drops across the ICV. Several considerations are being made to improve the results, especially outside the steady-state regime. The enhanced data filtering techniques implemented in the system helped manage "noisy" data. The analytical techniques described enhance digital capability in optimizing oilfield production through affordable flow-rate estimation for intelligent wells. Introduction Analytical techniques typically based on dynamic data driven models or well and physical models can be used as alternatives to physical metering. The model described in this paper is a well and process flow model that can be incorporated into the real-time well-by-well production surveillance and monitoring architecture of the digital asset. This model provides a real time estimate of key well performance indicators at a zonal and well level and can be used to continuously report daily production volumes on a well-by-well and zone-by-zone basis. In addition to the benefits outlined, the system increases the time between well tests and minimizes field visits. In the petroleum industry, chokes are deployed primarily at the well head for controlling flow rates from wells,to provide back pressure to a reservoir,to avoid formation damage,to maintain stable pressure downstream from the choke, andto dampen large pressure fluctuations. Intelligent wells provide the advantage of enabling the deployment of downhole chokes along individual production zones. These chokes provide the functions outlined above at a zone-production level. Intelligent completions also provide real-time pressure and temperature for each production zone that can be integrated with ICV choke information to estimate zone-production rates.