The cold-production-recovery process, also known as cold heavy-oil production with sand (CHOPS), is a method for enhancing primary heavy-oil production by aggressively producing sand. It is successful in vertical (or slanted or deviated) wells in western Canada. In this process, large amounts of sand are produced on a continuing basis along with heavy oil. Attempts at cold production in horizontal wells have not been particularly successful. When sand production has been generated in horizontal wells, these wells have tended to become plugged with sand.
This paper presents the results of experiments performed to assess the feasibility of applying cold heavy-oil production in horizontal wells that have been completed with slotted liners using less-aggressive (i.e., managed) sand-production strategies. Specifically, the effects of slot size, confining stress, fluid velocity, and sand-grain sorting on sand production were investigated.
The results indicate that slot-size selection is critical for establishing "sand on demand." From the experiments, a correlation between slot size and controlled sand production was found for well-sorted sands. This correlation should allow for the specification of appropriate slot sizes for target reservoirs containing well-sorted sands.
In the experiments, when flow rates resulted in low but persistent sand production, channels and/or elliptical dilated zones were created that greatly enhanced the effective permeability near the slot. This observation suggests that producing at low and steady sand cuts for a long period of time might bring two benefits: a way to transport the sand out of the well without causing plugging and the creation of high-permeability channels or zones that can improve production from the reservoir.
To summarize, if the appropriate slot size were combined with the right drawdown rates, controlled sand production could be achieved, with attendant significant increases in permeability. This suggests that substantially increased oil-production rates could be achieved from horizontal wells if sand-production rates could be maintained at low but persistent levels.
Rock mechanics tests on core from Early Cretaceous carbonate reservoirs from a super-giant field offshore Abu Dhabi has allowed definition of rock mechanical facies (RMF). Each of four RMF are based on stress-strain curves and associated strength and elastic parameters. The lab-based RMF correlate with mechanical stratigraphy classes previously defined from core (and that reflect visible differences in lithology and cementation). The RMF are correlated to reservoir zones and inter-reservoir, impermeable dense intervals, with three facies predominantly correlating with reservoir lithologies and one corresponding with primarily dense intervals. However, some reservoir zones, or sub-zones, can lie in more than one RMF. The RMF are, therefore, partly predictable: for any reservoir zone in the field prediction accuracy is to one or one of two RMF classes. This ambiguity is due to two factors: (i) lateral variation of RMF within some reservoir zones based on lithofacies; and (ii) continuity of mechanical properties between RMF classes. There is a change in RMF from crest to flank of the reservoir, as expected, but there is also local lateral variation within the crest of the field. The two RMF representing most of the reservoirs are expected to respond differently to field operations. Therefore, mapping lateral variation of RMF for some reservoir zones may provide a basis for implementing different reservoir management practices in different areas/zones of the field. The ultimate use of this information will be to enable full-field rock mechanics simulation of the reservoir to help understand the long-term effects of different production strategies.
Introduction & Background
The concept of mechanical stratigraphy is widely used, commonly to correlate fracture distribution and intensity to stratigraphy. The concept of rock mechanical facies (RMF) whereby a number of measured rock mechanical properties are correlated to stratigraphy is not new and is referred to in a number of papers, for example: Alhilali & Shanmugam (1991); Corbett & Friedman (1987); Yale & Jamieson (1994); McDermott et al., (2006); Khaksar, et al. (2009). However, RMF do not seem to be commonly used as a concept. We believe that characterising formations in terms of RMF has the potential to simplify characterisation for use for drilling; reservoir management; and history matching for simulation. In this paper we will describe how we have defined RMF for an oil-field and will discuss one way in which RMF could be used in the field.
The studied oil-field comprises a stack of limestone reservoirs separated by impermeable "dense?? limestone layers of Early Cretaceous age in a giant field located offshore Abu Dhabi (Figures 1 & 2). Production in the field has been by variably patterned water-flood over the last 30+ years. The dense layers measure up to a few tens of feet in thickness; the main reservoirs are up to 150 ft. The reservoirs are typically characterized by moderate to low matrix permeability, generally, but not exclusively, from 50 mD to 2 mD. Porosity is mostly in the range of 15-25%, more than half of which is microporosity. Depositional textures are predominatly wacke- to packstone with high-permeability streaks due to rudist and algal floatstone to rudstone and grainstones. Although intense bioturbation has destroyed most of the depositional textures, heterogeneities remain in some reservoirs in the form of dolomite-filled burrows, patchy/nodular cementation, stylolites and wispy solution seams, and fractures; all can occur as different layers within the reservoir. The reservoirs are not highly fractured although diffuse fractures are concentrated at the top and base of most reservoirs.
Introduction The most popular seismic attributes fall into three broad categories - those that are sensitive to lateral changes in waveform and structure such as coherence and curvature, those sensitive to thin bed tuning and stratigraphy, such as spectral components, and those sensitive to lithology and fluid properties - such as AVO and impedance inversion. We present a workflow that mimics multi-attribute clustering routinely done by human interpreters. This workflow automatically differentiates depositional packages characterized by subtle changes in the stratigraphic column as well as lateral changes in texture. Our 3D multi-attribute analysis is based on Kohonen selforganizing maps (SOM), which is one of the most commonly used unsupervised classification algorithms. The aim of this study is to map the heterogeneous Mississippian tripolitic chert reservoirs in a survey from Osage County, Oklahoma, using automatic seismic facies analysis.
Treadgold, Galen (Global Geophysical Services) | Eisenstadt, Gloria (Global Geophysical Services) | Maher, John (Global Geophysical Services) | Fuller, Joe (Global Geophysical Services) | Campbell, Bruce (Global Geophysical Services)
Brown, Morgan P. (Wave Imaging Technology Incorporated) | Higginbotham, Joseph H. (Wave Imaging Technology Incorporated) | Macesanu, Cosmin M. (Wave Imaging Technology Incorporated) | Ramirez, Oscar E. (Wave Imaging Technology Incorporated) | List, Dave (Fidelity E&P Company) | Lang, Chris (Fidelity E&P Company)
Mosher, Charles C. (ConocoPhillips) | Keskula, Erik (ConocoPhillips) | Kaplan, Sam T. (ConocoPhillips) | Keys, Robert G. (ConocoPhillips) | Li, Chengbo (ConocoPhillips) | Ata, Elias Z. (ConocoPhillips) | Morley, Larry C. (ConocoPhillips) | Brewer, Joel D. (ConocoPhillips) | Janiszewski, Frank D. (ConocoPhillips) | Eick, Peter M. (ConocoPhillips) | Olson, Robert A. (ConocoPhillips) | Sood, Sanjay (ConocoPhillips)
Gonzales, Veronica Monica (Schlumberger) | Barham, Michael R. (Helis Oil & Gas Co LLC) | Lawless, Paul (Helis Oil & Gas Co LLC) | Cherian, Bilu Verghis (Schlumberger) | Mata, Domingo (Schlumberger) | Higgins, Shannon Marie (Schlumberger) | Alatrach, Samer (Schlumberger)
The recent growth in horizontal well technology has resulted in existing oil and gas vertical development plays to be evaluated for horizontal well applicability. As operators attmept to evaluate the criteria for converting from vertical well plays to horizontal well plays, sound data gathering and modeling become crucial to understand how completion strategies needs to be modified for improved production, without utilizing an expensive trial and error methodology.
The Powder River basin contains a variety of producing shales and sands currently being explored for vialibility (i.e. Niobrara, Frontier, etc). In this study, reservoir and fracture properties are estimated based on hydraulic fracture modeling, rate-transient analysis techniques and production history matching to calibrate log data measurements. The challenges associated with calibration and modeling measurements from petrophysical and rock mechanics models are compared with hydraulic fracture and production modeling results to understand the direction of optimization and future basin growth.
Past experiences are typically the basis for design and implementation of developing a new drilling and completion program. Interpretation of the hydraulic fracture behavior is often inferred from simple diagnostics, and as production ensues the repeatability for success or failure is often attributed to modifying the hydraulic fracturing program or geological influences, which is subject to inconsistency and qualitative introspection. Within this study a single well modeling approach is utilized to understand fracture geometry, correlate this with production history matching results, and distinguish production attribution from hydraulic fracture characteristics or reservoir properties. Exercising this workflow addresses challenges affiliated with modeling fracture propagation and production matching and the gap associated with horizontal well development in existing vertical plays.