Optimization of the Performance of Partially Completed Horizontal Wells A. Retnanto, SPE, and M.J. Economides, SPE, Texas A&M University, C.A. Ehlig-Economides, SPE, Schlumberger, and T.P. Frick, SPE, Mining University Leoben
Goode and Wilkinson (1991) have presented an analytical solution for the performance of a partially completed horizontal well. In that work they suggested that normalized well performance (compared to the idealized open-hole equivalent) can be favorably disproportional to the fraction of open well, especially if the total open length is distributed among several spaced intervals.
This work, using a comprehensive multi- and single-well semi-analytical well performance model, extends the Goode and Wilkinson results examining various open lengths, reservoir thicknesses and permeability anisotropy ratios. Both transient and late-time (pseudosteady-state) results are shown. It appears that the number of segments where the normalized productivity index flattens out is about four for almost all thicknesses. For thinner reservoirs the relative impact from the number of segments is more pronounced. For example, for a thin (20 ft) isotropic reservoir, 40% of open well, of 2000 ft total length, distributed in four segments, would lead to almost 90% of the open-hole equivalent productivity index. For thicker reservoirs and/or longer wells the incremental productivity index fraction decreases. The idea of a sequence of open/close segments, accomplished either by perforated and non-perforated or slotted/blank liners has merit for zonal isolation, future well management and problem remediation. Economic calculations, incorporating production and well completion including zonal isolation devices show substantial attractiveness from partially completed, segmented horizontal wells.
Open-hole completion has been the one most commonly used since the early years of horizontal wells. It is still in wide use today although several other completion options are available.
The main reason for alternative well completions is that open holes do not allow flexibility for zonal isolation and future well management. Figure 1 illustrates the most common completion designs for horizontal wells. They are:
Open hole (Fig. 1a)
Slotted, perforated or pre-packed liner in an open-hole (Fig. lb)
Liner with External Casing Packers (Fig. 1c)
Cased, cemented and perforated (Fig. 1d)
Substantial work has been done on how to complete horizontal wells successfully, Reiss, Lessi and Spreux and Spreux et al. have presented completion technologies, especially as applied to selective completions.
The competence of the formation rock is a first consideration in deciding how to complete a horizontal well. Borehole instability and borehole collapse indicate shear failure, which in a horizontal wellbore is often severe, affected by high insitu stress anisotropy and excessive wellbore cooling. In an unconsolidated formation, sand production often becomes a problem.
This is an obvious reason against open-hole completions and in favor of, at least, slotted or perforated liners, sometimes augmented by sand-control screens. However, other considerations often may render liners as unattractive as open holes. Zonal isolation for production and fluid placement during stimulation may suggest the segmentation of a horizontal well with external casing packers (ECP) or other devices. To detect the origin of problems, such as water or gas influx, blank liners may be interspersed with slotted or perforated liners, all separated by external casing packers. In the case that a problem is detected through production logs, the zone may be shut off with a variety of means. Of course in the case of the ideal cemented and perforated well, problematic zones can be shut off with remedial cementing.