SPE Journal
Volume 17, Number 3, September 2012, pp. 727-741

Summary

Deconvolution is a technique for converting the pressure and rate data obtained from a well operating under variable-rate conditions into a much simpler form of constant-rate drawdown-pressure response function from the same well. However, deconvolution should only be applied to pressure and rate data that result from fluid flow in the reservoir that is governed by a linear set of equations. In gas reservoirs, the fluid-flow problem is nonlinear because the gas properties are strongly dependent on pressure. In some specific situations, this gas-flow problem can be linearized by the use of a pseudopressure transform that allows deconvolution to be successfully applied to those data. However, there are situations in which a pseudopressure transform alone does not fully linearize the gas-flow problem, and the resulting data will exhibit nonlinear behavior caused by the pressure dependence of the gas compressibility. Direct application of deconvolution to such data will produce erroneous results.

This paper presents an enhancement of the deconvolution algorithm that allows it to be used with pressure data affected by a nonlinear pressure-dependent compressibility. We discuss the detailed considerations for this enhancement of the deconvolution algorithm and demonstrate its performance on simulated test data exhibiting this nonlinear behavior. We also demonstrate the application of this enhanced deconvolution algorithm to a set of pressure data from a gas well fitted with a permanent gauge and covering the first 3 years of production from that well.

© 2012. Society of Petroleum Engineers

View full textPDF ( 1,441 KB )

History

• Original manuscript received: 20 October 2010
• Meeting paper published: 21 September 2010
• Revised manuscript received: 25 August 2011
• Manuscript approved: 14 September 2011
• Published online: 22 May 2012
• Version of record: 12 September 2012