Multiphase Well Surveillance with a Permanent Downhole Flowmeter

Abstract

Abstract Flow rate and fluid type (phase) are two of the most fundamental parameters needed in characterizing well performance. Traditional methods of estimating these parameters, particularly for real-time detection and diagnosis of production anomalies, have been limited by sampling frequency and data quality. This paper presents field test results of a new type of downhole multiphase flowmeter, which confirm the value of permanent downhole metering. The meter contains only three sensors but is capable of direct multiphase flow rate and cut measurements without slip models, even in highly deviated, recirculating flow. The physics basis and flow loop tests are discussed. Introduction Well monitoring, surveillance, and problem diagnosis are critical parts of the production business and many production parameters are monitored in the process. Of these, flow rate and fluid type (phase) are two of the most fundamental of measurements. Over the years, many instruments have been used to collect and process flow data, including production logging tools, surface test separators, and surface multiphase flowmeters, but none of these provides a complete information solution. Production logs provide flow information as a function of depth, but only intermittently in time. In addition, production logging tools are complicated, especially those designed for deviations beyond 45° from vertical, where sophisticated hardware with arrays of sensors must be combined with empirical slip models to cope with nonuniform and unsteady flows. Nevertheless, production logging tools are routinely used to update reservoir models and diagnose problems. In some cases, problems are discovered when using these tools that would never be diagnosed on the basis of surface measurements.1 A traditional method of flow analysis relies on routine periodic production testing through a separator and back allocation of production over the intervals between tests. Restricted access to a test separator often imposes constraints on when this information can be gathered, and the empirical relationships used to estimate rate between valid tests is often hampered by errors and uncertainties associated with varying flow conditions and data limitations. Finally, surface multiphase flowmeters are a good choice when the gas/oil ratio (GOR) is not too high and when there is room at the wellhead to install them. But in many instances, surface flowmeters have proved problematic because of the high volume of associated gas that evolves from the liquid stream as it flows up the well, complicating the determination of oil and water volumes. On platforms servicing many wells, there may not be enough room to install a meter on each well, thus preventing continuous flow rate and cut measurements on each well. This introduces the same type of uncertainty because of varying flow conditions as found in the test separator method. An alternative approach is to monitor the oil and water phases as deeply as possible in the well, minimizing the gas volume fractions. Historically, downhole multiphase flowmeters have relied on a simple Venturi device combined with a pressure gauge set 100 to 300 ft true vertical depth (TVD) above the Venturi.2 The pressure difference between the Venturi and the upper gauge is sensitive to the hydrostatic weight of the fluid column and its density and phase volume-fraction constitution. The combination of density and differential pressure across the Venturi provides total mass flow rate. There are two difficulties with this type of meter. First, in high-angle or horizontal wells, or where there are multiple reservoir zones, there is insufficient hydrostatic head for a practical density reading. Second, the simple hydrostatic density measurement ignores the effect of slip between the phases. Slip between phases creates very complicated flow patterns in high-angle and horizontal wells and is the reason for the complexity in production logging tools designed to operate in this environment.3 Empirical slip models can help fill the gap in two-phase oil and water flows, but are of limited use in three-phase flow. One reason is that gas density downhole can be orders of magnitude higher than at surface conditions and the slip laws are defined, in general, on the basis of data taken at surface conditions.