Deconvolution of Multiwell Test Data
- Michael M. Levitan (BP America)
- Document ID
- Society of Petroleum Engineers
- SPE Journal
- Publication Date
- December 2007
- Document Type
- Journal Paper
- 420 - 428
- 2007. Society of Petroleum Engineers
- 5.5 Reservoir Simulation, 5.6.3 Pressure Transient Testing, 5.6.4 Drillstem/Well Testing, 5.1 Reservoir Characterisation, 3.3 Well & Reservoir Surveillance and Monitoring
- 6 in the last 30 days
- 969 since 2007
- Show more detail
- View rights & permissions
|SPE Member Price:||USD 5.00|
|SPE Non-Member Price:||USD 35.00|
The deconvolution analysis technique that evolved with development of the deconvolution algorithms by von Schroeter et al. (2004), Levitan (2005), and Levitan et al. (2006) became a useful addition to the suite of techniques used in well-test analysis. This deconvolution algorithm, however, is limited to the pressure and rate data that originate from a single active well on the structure. It is ideally suited for analysis of the data from exploration and appraisal well tests. The previously mentioned deconvolution algorithm can not be used with the data that are acquired during startup and early field development that normally involve several producing wells.
The paper describes a generalization of deconvolution to multiwell pressure and rate data. Several approaches and ideas for multiwell deconvolution are investigated and evaluated. The paper presents the results of this investigation and demonstrates performance of the deconvolution algorithm on synthetic multiwell test data.
Pressure-rate deconvolution is a way of reconstructing the characteristic pressure transient behavior of a reservoir-well system hidden in the test data by well-rate variation during a test. The deconvolution analysis technique that evolved with development of the deconvolution algorithms by von Schroeter et al. (2004), Levitan (2005), and Levitan et al. (2006) became a useful addition to the suite of techniques used in well-test analysis. It has been implemented in commercial well-test analysis software and is routinely used for analysis of well tests.
This deconvolution algorithm, however, is applicable only for the case when there is just one active well in the reservoir. It is ideally suited for analysis of exploration and appraisal well tests. The previously described deconvolution algorithm cannot be used for well-test analysis when there are several active wells operating in the field and the bottomhole pressure measured in one well during a well test is affected by the production from other wells operating in the same reservoir. The deconvolution algorithm has to be generalized so that it is possible to remove not only the effects of rate variation of the well itself but also the pressure interferences with other wells in the reservoir. As a result, we would be able to reconstruct the true characteristic well-pressure responses to unit-rate production of each producing well in the reservoir. These responses reflect the reservoir and well properties and could be used for recovering these properties by the techniques of pressure-transient analysis.
Multiwell deconvolution thus becomes in a way a general technique for interference well-test analysis. The problem, however, is that the interference pressure signals produced by other wells are small compared to the pressure signal caused by the production of the well itself. These pressure interference signals are delayed in time and the time delay depends on the distance between respective wells. All this makes multiwell deconvolution an extremely difficult problem.
|File Size||1 MB||Number of Pages||9|
Levitan M.M. 2005. PracticalApplication of Pressure/Rate Deconvolution to Analysis of Real Well Tests.SPEREE 8 (2): 113-121. SPE-84290-PA. DOI: 10.2118/84290-PA.
Levitan, M.M., Crawford G.E., and Hardwick, A. 2006. Practical Considerations forPressure-Rate Deconvolution of Well-Test Data. SPEJ 11 (1):35-47. SPE-90680-PA. DOI: 10.2118/90680-PA.
Von Schroeter, T., Hollaender, F., and Gringarten, A.C. 2004. Deconvolution of Well-Test Data as aNonlinear Total Least-Squares Problem. SPEJ 9 (4): 375-390.SPE-77688-PA. DOI: 10.2118/77688-PA.