|Publisher||International Petroleum Technology Conference||Language||English|
|Content Type||Conference Paper|
|Title||Analysis of Flow Mechanisms in a Faulted Carbonate Oilfield through Reservoir Simulation and Experimental Design|
Gael Gibert and Francois Michel Gouth, Total E&P Qatar; and Rashed Noman and Abdulla Ahmad Al-Suwaidi, Qatar Petroleum
International Petroleum Technology Conference, 7-9 December 2009, Doha, Qatar
Copyright is retained by the author(s).
This reference is for an abstract only. A full paper was not submitted for this conference.
The limited ductility of carbonate rocks in general with regard to tensile or compressive stress results in a significant C_- if not predominant C_- proportion of carbonate oilfield showing static and/or dynamic evidence of fracturation. This natural fracturation can occur and have an impact on production mechanisms at different scale within the reservoir, from diffuse fractures to regional faults. Al Khalij field is located offshore Qatar in the Mishrif formation composed of Cenomanian limestones. The reservoir has a limited thickness and is made up of a layercake of high porosity/permeability oil bearing and low porosity/permeability water bearing layers affected by early and late diagenesis lying on an active aquifer. The field is essentially developed with long horizontal wells and peripheral water injection. Continuous reservoir monitoring acquisition, the resulting dynamic synthesis as well as borehole imagery logs and seismic imaging tend to bring into light the possible presence of dynamically active faults previously undetected that are deemed to link the aquifer with the reservoir and to result in observed water cut increases and production interferences. Decision has been taken to put a slant on locating and qualifying these events taking advantage of an integrated Geosciences and Reservoir study. On the one hand a comprehensive seismic and geological study as been carried out (See IPTC09 abstract proposal Fault detection workflow in a dimmed structural context: Al Khalij field in Qatar by B.Klein et al.); on the other hand authors dedicated themselves to describing complex flow mechanisms close to these faults in order to assess their impact in terms of oil production, reserve forecasting and relevance of current operating methods.
2. OBJECTIVES OF THE STUDY
In order to achieve the goal of the study - which is to precisely describe and understand the flow mechanism brought into play when a horizontal well is producing near a conductive fault - authors have chosen to drop milestone objectives which are: •Assess and sort relevant parameters and features governing fluid flows around horizontal wells and faults. •Reproduce fractional flow history of chosen typical wells where faults are deemed to come into play and explain related flow mechanisms. •Relate witnessed watering/dewatering interferences between two wells with the presence of faults and quantify their effect. •Assess the impact of the uncertainty/variability of previous parameters on fractional flow and interferences. In order to focus on flow mechanisms and their impact on production, the study has a limited scope: •This study does not include fault characterization using pressure transient. •This paper is not aimed at quantifying the productivity of horizontally fractured horizontal wells. •This paper does not focus on theorical fracture modeling in reservoir simulation models.
3.METHOD •Build a gridded sector model of a representative portion of the field with a limited number of well and faults. •Screen parameters highlighted by the dynamic synthesis and rank them in terms on uncertainty, relevance, and impact on flow mechanisms and production. •Systematically explore the variability of flow mechanisms through Experimental Design and assess correlation between parameters. •Reproduce with the sector model the fractional flow and interferences level of selected wells thanks to selected parameters values; describe and explain occurring flow mechanisms.
4.RESULTS AND OBSERVATIONS •Effects of connection/disconnection between the well and a conductive fault has been assessed in terms of mechanisms and oil production. Thorough quantitative investigation of disconnection length has been conducted. •Interferences between two wells producing nearby the same conductive fault has been reproduced. Effects of the variation of selected parameters have been assessed. A particular impact of the distance between the two wells and their liquid production rate has been observed. •Complex water cut evolution has been matched for several typical producing wells thanks to the variations of relevant parameters. Effects of specific reservoir features and parameters has been highlighted, explained and linked to the nearby presence of a conductive fault. Effects of the fault on the local formation pressure have been documented.
5.CONCLUSIONS •Production through conductive fault has been deemed to play a key part in the production mechanism of currently producing wells. Variation of behavior from well to well can be explained by different fault setup near the well, different degree of connection to the fault and variability of reservoir parameters. •Observed formation pressure gradient induced in produced layers by faults that are not connected to wells has been modeled and reproduced, validating this phenomenon as a source of indirect oil sweeping by water aquifer. •The different interferences seen on water cut evolution of certain wells are explained by the presence of a conductive fault in the proximate vicinity of these wells. Both detrimental (WCT increase) and beneficial (WCT decrease) can appear depending on the distance between the wells, their liquid production rate and the degree of connection to the fault. •The degree of disconnection of a well to the nearest fault (in terms of distance between the fault and the closest perforation set) plays the major role in the production mechanisms. Moving perforation set away from a conductive fault results in better oil production rate. From a certain disconnection length onward, the well performance increase is less important and tends to stabilize. This implies that there is a minimum value of disconnection length to achieve in order to maximize the well oil production.
6.APPLICATIONS Reservoir Understanding: •Detect the presence of faults near well bores thanks to production data •Detect the presence of faults with abnormal pressures seen in WFT measurements. •Allow a posteriori assessment of fracturation setup near horizontal wells thanks to their historical production data and as a result characterize a global fault network. •De-correlate matrix-driven and fault-driven behavior of a producing well. Derive matrix and fault properties. •Locate non-drained oil due to low areal sweeping efficiency. Reservoir Forecast: •Base for initialization and history matching of a full field reservoir model •Maximize field potential and achieve better Reservoir Management thanks to interferences understanding. •Drill new wells (infills and step-out) according to the fault pattern/spacing emphasized by the previous assessments. •Forecast horizontal wells performance and long term behavior taking into account possible detrimental effects of interferences. Operations: •Assess production mechanisms of currently producing well and propose innovative work overs to prevent production from faults. •Design of fit-for-purpose monitoring actions intended to locate and qualify faults (Open Hole and Cased Hole). •Contribute to design a new perforation procedure: assess the minimum acceptable distance between a perforation set and a possible conductive fault seen on logs or during drilling and position swell packers to limit possible water inflows. •Contribute to optimize the injection pattern taking into account the indirect oil sweeping due to pressurized faults.
7.TECHNICAL CONTRIBUTIONS •Modeling of complex polyphasic fluid flows around conductive faults and horizontal wells with a phenomenological approach. •Systematic investigation of the impact of a dozen of uncertain parameters through Experimental Design. The study can thus be seen as a real numerical experiment.
|File Size||52 KB||3|