Abstract

Frequently, it is seen that production rate and flowing tubing pressure decline rapidly on abnormally pressured wells following a hydraulic fracture pressured wells following a hydraulic fracture stimulation. It is often thought that rapid decline in production rate is due to inadequate fracture production rate is due to inadequate fracture characteristics. However, the rate of decline depends on many factors, such as formation permeability, fracture half-length, fracture conductivity, reservoir pressure and reservoir size. Also on deep, low-permeability, high-pressure wells, closure stress can play an important role on the rate of decline in production when large rapid pressure drawdown is required for economic production rates.

This paper utilizes pressure transient test data and production data to evaluate a hydraulically fractured well in a geopressured reservoir. A prefracture buildup test, type-curve techniques for fractured wells with boundary effect, and a two-dimensional, single-phase simulation program with closure stress on proppant are used to arrive at an explanation for the proppant are used to arrive at an explanation for the performance of a well. It was found that a small performance of a well. It was found that a small reservoir drainage area and a loss of fracture conductivity were major factors of rapid decline in productivity. productivity. Projections on future well performance were made and compared with subsequent actual well performance.

Introduction

To predict long-term performance for wells stimulated by massive hydraulic fracturing, reliable estimates of fracture half-length, fracture conductivity and formation permeability are needed. A deep geopressured formation in south Texas had been hydraulically fractured with subsequent rapid decline in production and flowing pressure. An evaluation of the production and flowing pressure. An evaluation of the fracture treatment was desired to determine if the loss of fracture conductivity due to closure stress on the proppant, or inadequate fracture length, was the cause of the decline. If either of these were the case, the well would be a candidate for a refracturing treatment. Prefracture pressure buildup data were available and post-fracture pressure buildup and postfracture production were used with type curves and a postfracture production were used with type curves and a reservoir simulator in evaluating and arriving at conclusions.

PREFRACTURE BUILDUP PREFRACTURE BUILDUP The purpose of a well-test interpretation is to identify the system and other reservoir characteristics, such as formation permeability (k), van Everdingen-Hurst skin factor(s), initial conditions or boundary of the drainage area, etc. Several papers cover the well-test interpretation for unfractured wells.

The prefracture buildup on this well was conducted by measuring the shut-in surface pressure instead of the bottom-hole pressure due to the nature of abnormally high pressure. The bottom-hole pressures were calculated from the shut-in surface pressure by using the Cullender and Smith correlation which appears in Appendix D of Reference 15. The calculated bottom-hole pressures were then used in the interpretation. (It should be noticed that when bottom-hole pressures are mentioned, they are referred to as calculated bottom-hole pressures throughout this paper.)

The well was shut in for 90 hr; however, only 48 hr of the buildup data was usable for the analysis due to mechanical problems with the pressure gauge at very high pressure.

Figure 1 is a log-log plot of change in surface pressure vs shut-in time of the prefracture buildup. pressure vs shut-in time of the prefracture buildup. Figure 2 is a log-log plot of change in calculated bottom-hole pressure vs shut-in time for this same buildup. In this case, there was no difference in the interpretation between surface and bottom-hole conditions. However, the fluid properties at bottom-hole conditions must be used in the calculation of permeability. A unit slope of early-time data on the permeability. A unit slope of early-time data on the log-log plot indicates that wellbore storage is

P. 217

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