Interference Testing

Introduction

In most well tests, a measurement is made of the pressure response at the formation face of a well to a stimulus within that same well.  In the case of interference testing, the pressure response to a stimulus at one well is measured at another well.  The second well is relatively remote from the active well (the well where the stimulus is applied).

 

Interference tests are carried out to test for reservoir continuity, to detect directional permeability and other major reservoir heterogeneities and to estimate reservoir volume.

 

A pulse test is a modification of standard interference testing that is designed to assist in identification of the signal produced by the stimulus.

 

In an interference test, a long duration rate modification in one well creates a pressure interference in an observation well that can be analyzed for reservoir properties.

 

Interference and pulse testing require at least one active (producing or injecting) well and at least one pressure observation well.  The observation well is shut-in while the testing is performed.

 

Usually, both the mobility-thickness product, kh/m, and the porosity-compressibility-thickness product, fcth, can be estimated.

 

In reservoirs with fluid contacts in the zone of influence, results may be unreliable.

 

Single Injecting Well and a Passive Monitoring Well

Consider that there is an injector (the active well) and there is a monitoring well (the passive well).  Assume further that the monitoring well has never been produced from or injected into.  The theoretical response for measurement at the passive well is:

 

 

In oilfield units, this can be expressed as:

 

 

Two Well Interference with Flow/Injection

If there is a history of injection/production in both the active and the passive wells, the situation is more complex.  Assuming single phase, small compressibility liquid and a uniform, homogeneous, infinite, horizontal reservoir:

 

where:

 

d is the distance between the two wells.

 

Analysis

For infinite acting situations there is a set of available type curves (Mueller and Witherspoon, JPT, April 1965; Earlougher, 1977).  These are developed using the following nondimensionalization.

 

 

Deviations from fits on the type curves can be indicative of reservoir heterogeneity.  Finally, be aware that there is no real guarantee of a unique fit to the type curves.  Unless the reservoir is homogeneous and isotropic, the interpretation is ambiguous.

 

For bounded systems, dimensionless data have been published.  In either case, software packages can facilitate analysis.

 

If the wells are much closer together than the nearest boundary or to another active well (by a factor of ~10) and the stimulus is long lasting, it may be possible to use the logarithmic approximation.  Under these circumstances:

 

 

 

Plotting pws versus logt and taking the slope gives:

 

 

 

For one shut-in observation well and multiple producing/injecting wells around it, the superposed relationship for the observation well is:

 

where:

 

q .............. production/injection rate at the observation well before it was closed in,

qj ...................................................................... production/injection rate at Well j,

tj ......................... producing/injecting time at the jth interfering well prior to shut-in

                                                                                          of the observation well,

Dtj ................................... producing/injecting time interval of the jth interfering well

                                                                   since the observation well was shut-in,

n ......................................................................... number of interfering wells, and,

aj ............................... distance of the jth interfering well from the observation well.

Precautions

1.     Additional considerations are required if boundaries are close.

2.     The rates at the active wells should be fairly constant.

3.     WATCH your signs – Injection is negative.

4.     Wellbore storage and skin effects at the pulser and the responder (active and passive wells) can lead to complexities and uncertainties in the results, especially if the dimensionless distance between the two wells is less than 1,000 or if the duration of the pulse is too short.

5.     There are other complicating factors for pulse testing (lunar-solar tides, oilfield activities, natural or hydraulic fractures, multi-layering, the presence of gas ….Lunar and ocean tides are periodic.  In an inland reservoir, these can cause pressure variations of ±0.015 psi.  In near-shore and offshore reservoirs, the pressure fluctuations due to tides can be ±1 psi.  The high compressibility of gas tends to suppress or attenuate the pressure signal at the observation well.

6.     Unless, and even if, the reservoir is homogeneous, isotropic and infinite, interpretation can be ambiguous and conjectural.

 

References

1.     Daltaban, T.S., and Wall, C.G.: Fundamental and Applied Pressure Analysis, Imperial College Press, London (1998).

2.     Earlougher, R.C. Jr.: Advances in Well Test Analysis, Henry L. Doherty Memorial Fund of AIME, SPE, Dallas (1977).

3.     Matthews, C.S. and Russell, D.G.: Pressure Buildup and Flow Tests in Wells, Henry L. Doherty Memorial Fund of AIME, SPE, Dallas (1967).

4.     Sabet, M.A.: Well Test Analysis, Gulf Publishing Company, Houston (1991).



Back to Key Issues Previous Top Monitoring Best Practices