Abstract

The possible range of, and the application of standard single phase interpretation methods to, the pressure transient responses for hydraulically fractured wells producing gas-condensate fluids are examined in this paper. A fully implicit, EOS based, compositional model is used to generate buildup responses for various combinations of fracture half-lengths, fracture conductivities, fluid richness levels, rates, reservoir relative permeability curves and producing times. Though all the responses presented here assume that average reservoir pressure remains above the dewpoint of the initial fluid in situ, the sensitivity to the difference between the initial reservoir pressure and this dewpoint pressure is also investigated. The effects of Non-Darcy flow in the reservoir and or the fracture, wellbore storage and capillary or Bond number dependent relative permeabilities are not considered. In addition, most of our results assume that the fracture proppant has straight line relative permeability characteristics. Some limited information, however, is presented where this assumption is relaxed.

The results of this work indicated that these buildup responses can be separated into two basic categories depending on the specific values of the above parameters. In the first category, the derivative response shape is not noticeably different from that of a fractured well with fracture face skin. The second category is typified by a derivative shape dominated by a hump between the early fracture flow periods and the late pseudoradial flow periods. A method is presented for deriving reasonable estimates for true completion skin factors for responses in this category. The reliability of applying standard single-phase interpretation methods (straight line analyses and history matching) is documented for both categories of responses and error directions and magnitudes are given.

Introduction And Literature Review

The accumulation of a condensate ring around a well producing at a pressure below dewpoint can cause a significant loss in well productivity. A number of researchers have documented this productivity loss in both simulation studies and measured field data. See References 1-6 for some examples. Bourbiaux presents a parametric modeling study that investigates depletion behavior in gas condensate wells. Primary parameters considered in their study are fluid richness, relative permeability curves including curves for IFT and non-Darcy dependence and production constraints. Some attention is also focused on reservoir permeability. Carlson and Myer study the effects of condensate dropout on the performance of a fractured well in a lean condensate reservoir. They also present limited information on well test analysis in fractured gas condensate wells. Settari, et. al use a case study to discuss productivity of fractured gas condensate wells: parameters studied are vertical heterogeneity, fracture length and conductivity, partial perforation, and relative permeability curves (excluding IFT or non-Darcy curves). References 2 and 3 point out that productivity loss in an unstimulated well can be mitigated significantly by fracturing the well. Afidick, et. al also present a case study of a gas condensate reservoir. Fractured wells are not considered in their study; however, long term performance and some well test analysis in vertical wells with condensate dropout are presented. Vertical heterogeneities are also represented in their modeling studies. Jones, et. al present methods for interpreting buildup responses in gas condensate wells. They present a two phase analog that can be used for wells producing below the dew point pressure, that correlates well with a liquid solution. They show that shape of the buildup response is governed by the saturation profile at shut-in.

P. 695

This content is only available via PDF.
You can access this article if you purchase or spend a download.