Better Permeability Estimation From Wireline Formation Testing

You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.

To ensure continued access to JPT's content, please Sign In, JOIN SPE, or Subscribe to JPT

The use of pressure-transient data in formation testing to describe reservoirs is considered mature technology, particularly when applied to data collected through production testing. The extension of this technique to data obtained using wireline formation testers (WFTs) has been gaining momentum in the industry; however, the integration of these outputs with other measurements of data is not always straightforward. The complete paper presents different methods of using pressure-transient data from WFTs; many of these methods are summarized here.

Pressure-Transient Data From WFTs

Perhaps the most widely used form of WFT pressure-transient data is that derived from small-volume drawdowns and buildups during a pressure test. The volume of fluid withdrawn from the formation, and the resulting depth of the pressure pulse, is limited to the near-wellbore region. The flow regime that develops during these tests is typically spherical flow in an infinite medium; hence, the mobilities derived from these sorts of pressure-transient tests are spherical mobilities and need to be converted to radial mobilities to quantitatively compare the tests. Additionally, pretest-derived mobilities have two fundamental challenges: the unknown effect of skin caused by drilling damage and the uncertainty of fluid viscosity to be used to convert the resulting mobility to permeability.

The other common application of pressure-transient information during wireline-formation tests uses pressure data over a much longer interval. During an extended pumping station with a WFT, a particular flow-rate history is applied to a well and the resulting pressure changes are recorded. From the measured pressure response, and from predictions of how reservoir properties influence that response, an insight into the reservoir can be gained. In order to make these predictions, it is necessary to develop mathematical models of the physical behavior taking place in the reservoir. Fig. 1 shows the difference between the volume investigated with a small-volume pressure test and an extended pumpout station. The most-common well model that is used when interpreting WFT data is the vertical limited entry model.

Fig. 1—The flow regime that develops during a small-volume pressure test (left) and during a much longer pumpout station (right).

 

Fluid flow in porous media is governed by the diffusivity equation. To derive it in its simplest form, the following assumptions and simplifications have to be made:

  • The reservoir is homogeneous, isotropic, and of constant thickness
  • The flow is horizontal
  • The fluid is monophasic and slightly compressible
  • Pressure gradients are small, and Darcy’s law applies

Analyzing the log-log plot, in which the pressure rises during buildups and its derivatives are plotted against superposition time, allows the deriving of a skin-independent mobility-thickness estimate for the effective flowing unit tested. Although not trivial, the thickness of the flowing unit can then be used to derive the effective permeability of the flowing phase.

Using Pressure-Test-Derived Mobilities Quantitatively

Mobilities derived with small-volume pressure tests typically use the pressure drop associated with an applied flow rate through a device of known flowing area and the resulting buildup. The assumption is that the pressure drop observed is purely a result of the dynamic properties of the reservoir. In reality, small-volume pressure-test-derived mobilities and resulting permeabilities are affected by a number of factors, namely

  • Mechanical skin caused by drilling damage and fines migration in the near-wellbore region
  • Variation of viscosity of the fluid in the near-wellbore region and invaded zone
  • The mobility derived with small-volume pressure tests is generally spherical and, depending on the heterogeneity of the formation, the assumptions used to convert this to a radial mobility may or may not be reasonable

The effect of the drilling fluid used on resulting mobilities is significant. Wells drilled with water-based muds (WBM) or polymer muds appear to create higher and more-heterogeneous skin. The amount of solids in the drilling fluid used can also create additional skin by plugging up the pores in the near-wellbore region. Clean synthetic oil-based muds (OBM) with low solids content have been observed to create a more-uniform and -efficient mudcake, thereby reducing the variability of skin across a depth interval. The effect of different viscosities of the invading fluid and native formation fluid along with the miscibility of these two fluids is least pronounced in oil zones drilled with OBM and water zones drilled with WBM. As such, it is reasonable to attempt to use sufficiently numerous small-volume-test-derived mobilities to estimate the permeability quantitatively in a reasonably homogeneous light oil interval drilled with OBM.

Using Radial Permeability From WFT Pressure-Transient Analysis (PTA) for Quantitative Permeability

The primary differences between using small-volume tests and extended pumping stations for pressure-transient information are

  • The pressure pulse that is propagated extends well beyond the near-wellbore region, typically a few decameters into the formation.
  • The viscosity of the primary fluid being investigated is known.
  • Skin can be compensated for; therefore, the resulting mobility thickness from the interpretation is skin-independent.

With sufficient good-quality and repeatable pressure-buildup data, the interpretation results in a reliable estimate of the mobility thickness of the flow unit being investigated. Converting the mobility thickness to an average permeability requires the input of viscosity of the mobile phase of the native formation fluid and the thickness of the flowing unit.

Using Post-Sampling Pretests and WFT PTA To Identify Excessive Skin in Presampling Pretests

Formation damage caused by the drilling process can affect near-wellbore mobility. For a typical WFT pretest, drawdown volume ranges from a few to tens of cubic centimeters, so the investigation depth is mostly within the damaged zone near the wellbore. As a result, pretest mobility from WFT may underestimate formation mobility. WFT PTA requires flowing formation fluid at significantly larger volumes (tens to hundreds of ­liters) compared with the pretest, so the investigation depth of WFT PTA is typically beyond the near-wellbore damaged zone and reflects actual formation mobility. Additionally, flowing large volumes of formation fluid can potentially clear the damaged zone, so it is repeatedly observed that pretest mobility after flowing is higher than that before flowing, even after correction for fluid-­viscosity differences between the mud filtrate and the formation fluid. 

Using Radial Permeability From WFT PTA for Identifying Varying Facies Within a Well

An extension of the quantitative use of WFT PTA for permeability estimation is to identify different facies within a well where other sources of permeability measurements are not available, and a porosity/permeability relationship based on core data is used for each facies. This may act as a trigger to investigate the facies because of differences in WFT PTA vs. core-derived data, or the WFT program may actually be designed with this objective specifically in mind.

Upscaling Discreet WFT-PTA Stations to Well Scale

In complex reservoirs, where the producibility across a series of stacked sand bodies may be of interest, it is not possible to perform an adequate number of discreet WFT-PTA stations. Through robust anchoring of dynamic tests using WFTs and continuous static logs, however, it is possible to describe the productivity or effective permeability thickness of the flowing phase by obtaining a few discreet WFT-PTA tests. This allows comparison of WFT data with the results of a conventional well test [drillstem test (DST)] across a much larger interval covering several reservoir sand bodies without performing a WFT PTA in each of those sand bodies. The complete paper provides a work flow that can be used to compare upscaled WFT-PTA tests with the results of a conventional well test. The higher granularity of WFT data as compared with well-test data can also be used for inflow predictions.

Using WFT-PTA-Derived Permeabilities in Heterogeneous Formations

While it is almost always possible to match a WFT-PTA response, care should be taken when attempting to use permeabilities derived from these tests, particularly in heterogeneous formations such as highly heterogeneous carbonates. It is always important to integrate the results of WFT-PTA tests with other sources of logs. Although quantitative use of permeabilities derived from WFT PTA in modeling, for instance, may be challenging, WFT PTA still has powerful uses—for example, to determine permeability-thickness cutoffs for a flow-vs.-no-flow scenario.

Using WFT-PTA Buildups for Identifying Flow Boundaries

Caution has been exercised when attempting to define flow boundaries away from the wellbore using WFT-PTA tests. A caveat that is often associated with these tests is that one should rely on conventional well testing (DST) in order to define lateral flow boundaries. While this is true for defining the minimum size of a reservoir, oil and gas fields today are becoming more complex and compartments created by faults or isolated reservoir bodies are more common.

Using WFT-PTA Buildups for Identifying Permeability Anisotropy

Vertical interference tests using WFTs are performed by setting either a standard probe or dual packer as the flowing device, along with one or more observation probes a distance away from the flowing device.

Discussion and Conclusions

  • Real-time monitoring and control is an absolute necessity when attempting to obtain quality pressure-transient data by use of WFTs.
  • The integration of various data sources is critical.
  • WFT-PTA tests should be limited to environments where a single phase is flowing from the formation.
  • Sufficient WFT-PTA tests should be performed over a reservoir interval, particularly in complex environments.
  • Permeability estimates from small-volume tests should be used only under special circumstances quantitatively, and care should be taken in order to correct this data accordingly.
This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper IPTC 18914, “Extracting More From Wireline Formation Testing: Better Permeability Estimation,” by S.R. Ramaswami, P.W. Cornelisse, SPE, H. Elshahawi, M. Hows, and C.L. Dong, SPE, Shell, prepared for the 2016 International Petroleum Technology Conference, Bangkok, Thailand, 14–16 November. The paper has not been peer reviewed. Copyright 2017 International Petroleum Technology Conference. Reproduced by permission.

Better Permeability Estimation From Wireline Formation Testing

01 February 2018

Volume: 70 | Issue: 2

STAY CONNECTED

Don't miss the latest content delivered to your email box weekly. Sign up for the JPT newsletter.