Layer | Fill | Outline |
---|
Map layers
Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|
Fill | Stroke |
---|---|
Collaborating Authors
Results
Abstract A new processing workflow has been engineered to combine reservoir deliverability, defined by production logging (PL) measurements, with nodal analysis evaluation. This allows the effects of various completion modifications to be quantitatively modeled, predicting the resulting changes in well production prior to intervention. A graphical computer interface is used to implement the workflow and facilitate comparisons of the effect on production of multiple completion/recompletion scenarios, using "drag and drop" icons for placing simulated plugs, patches, perforations, etc. on the base profile. Beginning with the flow profile and formation pressure, the process defines zonal reservoir parameters in terms of the productivity index, water cut, and gas/oil ratio. These data are passed transparently to a nodal analysis package that automatically creates a model to match the measured rates and pressures as closely as possible to allow production computations. The process is designed to significantly decrease workover decision time and risk. This technique provides opportunities to compare and recommend formation evaluations, profile modifications, various artificial lift options, and stimulation activities to maximize production. The modified completion is then passed to the nodal analysis engine and a new predicted production profile is computed and displayed. Outputs include graphic and ASCII format log data, ASCII format tables, and data files, which can be used for further studies. In this paper, the use of the workflow is demonstrated by one example and two case studies, including a selective inflow performance analysis of a gas well and a successful water shutoff operation in an oil well. PL answers provide an opportunity to evaluate the downhole flow performance of multizone completions. Nodal analysis techniques provide a method to model the effects of the fluid properties and completion configuration as the fluids are produced to the surface. Combining PL answers seamlessly with nodal analysis provides unique insight into the reservoir completion performance and production improvement potential. Introduction A workflow will be discussed that permits the combination of PL data with nodal analysis. This method is used to rapidly screen worker scenarios to optimize the performance of the well when planning an intervention. The workflow, based on production log data, has been automated and computerized to speed up the planning of remedial wellbore operations and reduce the risk of failure of such interventions. The workflow user does not need advanced knowledge of nodal analysis techniques or software, because as the nodal model is set up automatically from the PL data and well sketch, which greatly speeds up the process. Being able to examine a range of wellbore intervention options using the workflow before any actual intervention reduces the risk of a workover failure and allows an optimal workover to be planned. The example workflow mentioned earlier demonstrates the technique and the two case studies present actual field usage. Production Logging Techniques Production logging is used to determine both single and multiphase fluid entry profiles in both producing wells and injectors. The primary information derived from PL data is a record of pressure, temperature, and flow rate versus depth. This lets us know which zones in the well are producing; what phases (oil, water, and gas) are being produced; and whether there are any nonproducing intervals. In shut-in wells, we can also look for crossflow between various reservoir layers.
- Europe > United Kingdom > England > Hampshire Basin > PL 089 > Block 98/6 > Wytch Farm Field > Sherwood Formation (0.99)
- Europe > United Kingdom > England > Hampshire Basin > PL 089 > Block 98/11 > Wytch Farm Field > Sherwood Formation (0.99)
- Europe > United Kingdom > England > Hampshire Basin > PL 089 > Block 97/15 > Wytch Farm Field > Sherwood Formation (0.99)
- (5 more...)
Abstract Conventional production log interpretation using a spinner assumes that the spinner responds to the holdup-weighted average velocity of the fluids traveling through the swept area of the spinner blades; however, the spinner tends to respond preferentially to the most dense phase flowing through it and so can be considered as a momentum-averaging sensor. This means that when fluids of markedly different density and velocity are averaged using the current best practice we may not be calculating the most accurate spinner velocity to be used in the production log interpretation. This paper examines alternative spinner averaging approaches and then evaluates the difference on synthetic, laboratory, and real downhole data sets. We conclude from this short study that there is potential to improve the current practice of spinner interpretation, especially in gas/liquid flows, and thus provide better data for gas production allocation and gas reservoir management. Further work is needed to develop this approach. Introduction Spinners are a form of ‘turbine’ flowmeter. Turbine flowmeters are used extensively in industry; they have excellent accuracy in nearly all applications. The only significant case in which the turbine flowmeter (spinner) uncertainty can routinely exceed 100% is in multiphase production logging (PL). Design constraints for downhole application mean that, unlike their industrial counterparts, the spinner never occupies the full pipe cross section. In addition, the spinner often uses a simple ‘flat-blade’ design so that it can collapse to a suitable size for running in and out of production tubing. Normally we assign these large spinner errors to recirculating flow, stratification of the flow, or even errors in the slip model. While these sources of error are undoubtedly present, we also have to consider the way in which the spinner is assumed to average the velocities passing through it. Let us briefly review the standard PL spinner model.[1] In monophasic flow the spinner sits centred in a pipe and samples the centre line velocity present in the pipe. The apparent velocity registered by the spinner overestimates the average velocity and is corrected by the velocity profile correction factor, FVPC, which is a function of the Reynolds number, pipe internal diameter, and spinner diameter. Formula 1 The assumptions in this procedure are only strictly valid in single-phase flow, but in the absence of any effective multiphase models, we have no practical alternative for multiphase flow. For multiphase gas/water flow there are three main approaches that can be used to relate the spinner-derived two-phase velocity to the individual velocities of the gas phase, VG, and the water phase, VW. The Volumetric Model The current practice in two-phase flow is to assume that the spinner measures the holdup-weighted average of the phase velocities present. Formula 2 To understand the implications of this paper, consider a hypothetical well with 50% water holdup and 50% gas holdup and with gas travelling at 2 m/s and water travelling at 1 m/s. The standard volumetric model for the spinner indicates that we should expect to derive a corrected spinner velocity of 1.5 m/s. But if the spinner responds to phase momentum, water, being denser, will have a much greater influence on the spinner blades than the relatively less dense gas. In this case, the corrected spinner velocity should be much closer to the heavy phase velocity of 1 m/s.
- Europe (0.69)
- North America > United States > Texas (0.28)