Permanent downhole gauges(PDGs) provide vast amounts of pressure-transient and rate data which may be interpreted with improved pressure-transient analysis (PTA) approaches to gain more knowledge about reservoir dynamics. Permanent pressure and rate measurements allow for analysis of time-lapse pressure transients and comparative interpretation of flowing and shut-in periods. The approaches used provided the basis for an improved methodology of interpreting permanent pressure measurements, where the scope of the standard PTA application may be extended to integrate new data sources
A Methodology of PDG Interpretation Focusing on Both Flowing and Shut-In Periods
Practical Remarks on Comparison of Pressure Transients and Choosing a Model.
Comparison of different pressure transients is usually carried out on the basis of plotting all the transients and derivatives on the same log-log plot.
In practice, a difference between time lapse shut-in pressure transients does not necessarily indicate change in well reservoir parameters because such a comparison is usually carried out for rate-normalized data with a chosen reference transient. The rate before the shut-in period of interest governs the pressure-transient location on the loglog plot. An approximate value may be attributed to this rate because of the averaging of flow data, while permanent rate measurements may help in reducing this uncertainty.
Flowing pressure transients are usually normalized subject to variable rates during the flowing period. This makes comparison of these pressure transients more reliable.
Pressure derivatives are more representative in this sense, because well history before and during a pressure transient is accounted for by use of the superposition principle commonly used for the derivative calculation.
At the same time, assuming radial flow as the main flow regime in the superposition calculation—as well as averaging, cutting, or possible errors in rate history before the pressure transient of interest—may have an impact on the derivative trend, especially for late elapsed times and interpreting boundary effects. Simulation of the well history in the linear scale with the analytical model used for the pressure and derivative interpretations in the log-log scale improves reliability of the analysis. The simulated pressure response may help in evaluating the impact of superposition effects and in revealing changes in well/reservoir parameters.
Comparison of time-lapse pressure transients and derivatives may be used for diagnostics of changes in well/reservoir parameters, while only simulation of the well history, or at least a part of the history, would provide reliable conclusions on such changes.
Choosing a proper model to describe the well, the reservoir, and boundaries is crucial for analysis and forecasting and for drawing conclusions. The usual practice is that the chosen model should represent basic well and reservoir features that are known before the analysis, such as well type; reference fluid and stimulation performed; and well environment, including neighboring wells and faults.
Comparison of Time-Lapse Pressure and Derivative Transients.
As a first stage of the analysis, time-lapse pressure and derivative transients may be extracted and plotted on the same loglog plot. Separate plots for well flowing and shut-in pressure transients are suggested because these two types of transients may follow different trends. Comparison of time-lapse transients may serve as a first indicator of changes in well/reservoir parameters.
Use of the superposition principle for derivative calculations in combination with possible impact of dynamic boundary effects (e.g., neighboring wells) may lead to deviation of the pressure and derivative transients from each other. This means that the comparative analysis of timelapse responses may be considered only as a preliminary diagnostic of changes in well/reservoir parameters.
Analysis of Time-Lapse Responses Focusing on Both Flowing and Shut-In Periods.
A second stage of the analysis is a step-by-step interpretation of each pressure transient or pair of transients (flowing and closest shut-in) according to historical data.
In general, a difference between flowing and shut-in responses may be related to many effects. Starting from the first analyzable period of the well history or, even better, from a flowing/shut-in pair, PTA is carried out in the log-log scale. First, the shut-in response may be used for matching pressure and derivative with a chosen model because this response is usually less noisy, with clear indications of flow regimes.
The model is further tested to be capable of matching the flowing response. Modifications of the model parameters, or even the model itself, may be necessary to fit the response.
If the chosen model provides a reasonable match of at least the flowing response, an attempt to match the well history (in the linear scale), or its segment, before the analyzed pressure transient may be made. Ideally, the model should provide a reasonable match of the whole well history. In practice, a reasonable match may often be achieved only for the limited history segment containing the pressure transient of interest, while changes in well/ reservoir parameters and presence of boundary effects varying with time may lead to deviations of the model response from the measurements.
Furthermore, the model is tested for capability of reproducing the next pressure transients chosen in the well history, and modifications of the model parameters and probably of the model itself may be performed. This step-by step interpretation would result in a set of well/reservoir parameters changing with time, providing the history of such changes and the current status of these parameters, which is of special interest for well-performance predictions.
Application of Interpretation Results.
The analytical or simplified numerical models applied in the interpretation process may be further used for prediction of well performance and behavior under different scenarios.
The models may be particularly useful for simulating short forecast scenarios, sensitivity studies, and uncertainty analyses. The time-lapse PTA also provides additional input for reservoir simulation.
Well-connectivity and reservoir properties are reported as time- or pressure dependent variables and may be used directly in reservoir models, improving history matching and prediction capabilities of the numerical models.
Field Cases Three field cases are presented in the complete paper to illustrate application of the methodology. The first case is a good example of classical PTA working well, providing reliable estimation of well/reservoir parameters from a singlewell shut-in or flowing response. The second example shows the value of interpreting both flowing and shut-in responses, with the advantage of using multiple shut-in pressure transients. The third, and most complicated, example illustrates all sides and advantages of the methodology applied.
Conclusions
Advantages of the time-lapse PTA were confirmed with field examples, providing estimation of well/reservoir parameters evolving with time, improving reservoir description through focusing on flowing reservoir properties, and understanding the difference between near-well-flow and boundary effects during flowing and shut-in periods. Application of the described methodology improves the PTA reliability and extends its scope in the following ways:
（1）Analysis of sequential pressure transients provides a basis for isolating reservoir effects from measurement noise.
（2）Comparison and interpretation of both flowing and shut-in pressure responses would give a more-complete picture of the well behavior with estimates of flowing well/reservoir parameters.
（3）Representation of well/reservoir parameters evolving with time is now feasible through available PDG data and interpretation approaches. As practical guidelines for timelapse PTA applications, the following may be suggested:
1）Use of both well-flowing and shut-in pressure transients.
2）Analyzing sequential flowing/ shut-in periods in well history to confirm repeatability (static well and reservoir conditions) or to reveal changes in well/ reservoir parameters.
3）Matching both pressure transient responses (the loglog scale) and history or its segments (the linear scale).
石油圈
