Authors: Lei Li, Carlos F Lange, Yongsheng Ma
Computational fluid dynamics has been extensively used for fluid flow simulation and thus guiding the flow control device design. However, computational fluid dynamics simulation requires explicit geometry input and complicated solver setup, which is a barrier in case of the cyclic computer-aided design/computational fluid dynamics integrated design process. Tedious human interventions are inevitable to make up the gap. To fix this issue, this work proposed a theoretical framework where the computational fluid dynamics solver setup can be intelligently assisted by the simulation intent capture. Two feature concepts, the fluid physics feature and the dynamic physics feature, have been defined to support the simulation intent capture. A prototype has been developed for the computer-aided design/computational fluid dynamics integrated design implementation without the need of human intervention, where the design intent and computational fluid dynamics simulation intent are associated seamlessly. An outflow control device used in the steam-assisted gravity drainage process is studied using this prototype, and the target performance of the device is effectively optimized.
Download paper
Authors: Marty Lastiwka (Suncor Energy) | Chris Bailey (Suncor Energy) | Bruce James (Suncor Energy) | Da Zhu (RGL Reservoir Management)
Over the past few years, an increasing number of operators in steam assisted gravity drainage (SAGD) in situ recovery of bitumen in the Alberta Oil Sands are becoming interested in the use of flow control devices (FCDs). Initial field trials by some operators of these devices have shown promise in improving steam chamber conformance, reducing incidences of steam breakthrough, high vapour production, and in addressing liner reliability concerns related to steam jetting.
While the application of FCDs is well-established in the conventional oil and gas industry to control gas and water coning, there are still a number of questions on how to implement FCDs optimally in SAGD. One major difference in the application of FCDs in SAGD compared to the conventional oil and gas industry is the high temperature environment with steam and elevated erosion risk.
The purpose of this paper is to present some practical considerations for the selection of FCDs and optimal completion FCD design for SAGD applications. In the first section, a discussion is presented on how to compare the performance of different flow control devices. Most devices have not been tested for SAGD, and there is a need for more comprehensive testing. The focus of the second section is on practical considerations for the installation of FCDs in a SAGD injection and production wells.
Download paper
Authors: M. Mahmoudi (RGL Reservoir Management Inc.) | S. Nejadi (University of Alberta) | M. Roostaei (University of Alberta) | J. Olsen (University of Alberta) | V. Fattahpour (RGL Reservoir Management Inc.) | C. F. Lange (University of Alberta) | D. Zhu (RGL Reservoir Management Inc.) | B. Fermaniuk (RGL Reservoir Management Inc.) | A. Nouri (University of Alberta)
The term skin is used to describe pressure drop caused by a flow restriction near the wellbore. The skin factor of wells completed using slotted liners can be explained by a number of phenomena including: the flow across the slots, flow convergence towards slots, near wellbore permeability, and occlusion of liner open area due to corrosion and scale deposition. This paper introduces an analytical skin model for the slotted liner, which incorporates these phenomena, and can be used to optimize the slotted liner design. The introduced analytical model was verified by physical and Computational Fluid Dynamics (CFD) models.
The proposed analytical skin factor model for slotted liners is based on slot width, slot density, the spatial distribution of slots, and near-liner permeability. The model also incorporates partial plugging of slots. The model is validated using experimental Sand Retention Testing (SRT) data. A series of SRT experiments were conducted at different flow rates for two Particle Size Distributions (PSD) from the McMurray Formation in Northern Alberta. The experiments were also modeled by the CFD to better understand the flow dynamic near the liner.
Results of the analytical model and experimental tests were generally in agreement. However, results of the analytical model deviate from experimental tests for narrow slots and high flow rates. In these cases, the analytical model predicts smaller skin than the experimental tests. For cases related to narrow slots and higher velocity the pore plugging close to the liner is significant which was not modeled in the analytical model. Moreover, for very fine sand (low permeability) sand-pack the deviation from the experimental results is higher in comparison with medium uniform sand (higher permeability) sand-pack. CFD simulations showed the effect of the slot width on the depth of the convergence zone, which is not included in the analytical model. Since the analytical model follows the experimental results for common flow rates in thermal production, the model could be used to assess the skin for different possible designs and choose the best slot specifications that minimize the skin.
This paper presents the details of an analytical model for the skin factor verified by experimental data and CFD simulation. This analytical model can be used to optimize the liner specification for the best flow performance. This paper also outlines the limitations of the analytical models for calculation the skin/pressure drop.
Download paper
Authors: Vahidoddin Fattahpour (University of Alberta) | Vittor Maciel (University of Alberta) | Mahdi Mahmoudi (University of Alberta) | Ken Chen (University of Alberta) | Alireza Nouri (University of Alberta) | Michael Leitch (RGL Reservoir Management)
There is a growing interest in physical model testing of the reservoir and large-scale sand control testing for oil sands. These experiments require the synthesization of representative sand-packs. Particle size distributions (PSDs) of these sand-packs ought to be comparable to the PSD of target oil sands. For practical and economic reasons, it is favorable to test samples with a limited number of PSDs, yet representative of a spectrum of oil sands. The aim of this paper is to categorize the PSD of Alberta oil sands to a limited but representative number for use in laboratory research.
This paper is based on the analysis of 152 PSD curves for Alberta oil sands. To categorize these PSD’s in a meaningful way, an algorithmic approach is presented which uses attributes that are widely used in sand control design (e.g. D10, D50, D70, fines content) and, subsequently screens and sorts the data to produce a finite number of PSD categories which represent the majority of the data. Rules are implemented in the algorithm to limit the number of categories (≤7), and require that each category cover a significant subset of the total data (≥10%).
A review of the published PSDs for oil sands across Alberta indicates a significant variation in the PSD curves even within the same reservoir. However, in spite of the fact that PSD data show a large variation, PSD categories can be identified to build representative oil sand samples for design and testing purposes. For the database used in this investigation, four major and two minor PSD classes were identified. These six PSD classes, cover more than 87% of the analyzed PSDs. Introduced classes and existing PSD classifications in the literature share interesting similarities. However, certain differences, such as the lack of very coarse ranges (D50~500 µm) was observed.
The method which is introduced for oil sand classification is based on the D-values which are commonly used in screen aperture design. This method provides a useful tool for both screen designers and researchers to categorize and focus their work on a specific set of representative PSDs, rather than a wide distribution of PSDs.
Download paper
Authors: Mahdi Mahmoudi (University of Alberta) | Vahidoddin Fattahpour (University of Alberta) | Alireza Nouri (University of Alberta) | Michael Leitch (RGL Reservoir Management)
This paper presents the results of an experimental investigation to determine the mechanisms of pore plugging and permeability reduction near SAGD screen liners. The aim is to arrive at a liner design that maximizes wellbore productivity without compromising the sand control function of the liner.
We set up a large-scale Sand Retention Testing (SRT) facility that accommodates a multi-slot liner coupon at the base of a sand-pack with representative grain shape and particle size distribution (PSD) of typical oil sands. Brine is injected at different flow rates and pressure differences across the coupon and the sand-pack as well as the mass and PSD of the produced sand and fines are measured during the test. Further, the PSD and concentration of migrated fines (<44 microns) along the sand-pack are determined in a post-mortem analysis. The testing results are used to assess the effect of slot size and slot density on the sand control performance as well as pore-plugging and permeability alterations near the sand-control liner.
We observed that the slot size, slot density and flow rate highly affect the concentration and PSD of produced fines as well as accumulated fines (pore clogging) above the screen. For the same flow rates and total injected pore volume, wider screen aperture and higher slot density result in lower fines accumulation above the screen but more sanding. Further, the variation of slot density alters the flow convergence behind the slots, hence, the size and concentration of mobilized fines. Results indicate that higher fines concentration near the screen reduces the retained permeability, hence, lowers the wellbore productivity.
This paper provides a new insight into pore plugging and fines migration adjacent the sand control liner. It also introduces a new testing method to optimize the design of sand control liners for minimum productivity impairment in SAGD projects.
Download paper
Authors: Matthew Miersma
One of the main methods of extracting oil from deep oil sands deposits is through the use of steam assisted gravity drainage (SAGD). For the best performance, inflow control devices (ICDs) are implemented along the SAGD production well to even out production and restrict unwanted fluids. Current methods of evaluating these devices rely on criteria that are dependent on the flow rate and fluid properties at which they are measured. In this study, three new criteria are proposed to evaluate and compare ICDs. These new criteria are derived from the physics of the flow in order to tie them to specific aspects of the flow and to have a reduced dependence on the flow rate and fluid properties. To further reduce the dependence of the criteria, they are calculated from a range of data, using a least squares fit. In order to evaluate the proposed criteria, detailed CFD models are developed for six fundamental ICD designs: a 15◦ nozzle, a 40◦ nozzle, a long channel, an expanding nozzle, a device based on Tesla’s fluidic diode, and a vortex based device. The CFD models are carefully tested to ensure they accurately model the flow. Using these simulations, the three criteria are calculated for each device. The criteria are then compared to the flow results and examined for flow and viscosity independence. Finally, the criteria are used to compare the six ICDs and identify the best design. The new criteria are not only better than existing criteria for comparing ICDs, but they are also specially adapted to support design development and optimization of new devices.
Download paper
Authors: Li, Lei & Ma, Yongsheng & Lange, Carlos
The complexity in configuring the CFD solver imposes a barrier for users to efficiently setup the solver and obtain satisfactory results. Such kind of deficiency becomes more obvious when it comes to simulation-based design where the CFD solver is expected to respond to design changes automatically. By applying artificial intelligence, expert systems can be used to capture the knowledge involved in CFD simulation and then assist the solver configuration. This paper proposes an expert system for both dry and wet steam simulation. According to the product design, the expert system is able to select the right module to model the steam flow. Based on the derived non-dimensional numbers, appropriate physics models can be selected to run the simulation. Grid adaption, higher order schemes, and a subroutine for advanced turbulence models help to improve the accuracy of the CFD model after rounds of simulation. The output of the expert system is a robust simulation model with accurate results which are guaranteed by flow regime validation, grid independence analysis, and error estimation. The effectiveness of the proposed system is demonstrated by the analysis of a contracted pipe. In dry steam simulation scenario, the error induced by the expert system is smaller than that of the traditional ANSYS batch mode. The results obtained by the expert system also match well the empirical results when it comes to wet steam simulation.
Download paper
Authors: Da Zhu (RGL Reservoir Management Inc.) | Ian D. Gates (University of Calgary)
Given the high viscosity of the oil, bitumen from oil sands reservoirs in western Canada is recovered by using steam which, due to its temperature, lowers its viscosity. One of the key issues faced by the operators is the steam conformance of the depletion chamber around wells. The greater the fingering phenomena of steam at the edge of chamber, the worse is the chamber uniformity and utilization of the well, and the greater are the green house gas emissions and water use per unit oil recovered. Fingering has long been explained as the penetration of steam phase into the oil phase which arises from an unfavourable mobility ratio. In this paper, we introduce linear instability analyses (Orr-Sommerfeld and Rayleigh-Taylor/Saffman-Taylor instability) of the interface between steam and oil layers and conduct a series of numerical simulations to reveal that fingering in the steam-assisted heavy oil recovery at the top of the steam chamber is created due to solution gas exsolution whereas fingering at the chamber edge is due to viscous shear instability. The results show that non-ideal steam conformance is inevitable even in homogeneous reservoirs.
Download paper
Authors: Lei Li, C. F. Lange, Yongsheng Ma
Multi-view feature modelling provides a specific view for each phase in product development. The analysis view should be fully integrated with CAD models in a multi-view product development environment for simulation-based design. In the development of fluid flow products, CFD (Computational Fluid Dynamics) is increasingly used as an advanced support. However, the successful application of CFD requires special knowledge and rich experience, which is a barrier for the conversion from the design view to the analysis view, and the maintenance of information consistency. Several approaches to multiple feature views have been proposed, such as design by features, feature recognition and feature conversion. In one-way feature conversion, features in a specific view are usually derived from the original design view. Bronsvoort and Noort put forward a multiple-way approach which enables a designer to modify the product model from an arbitrary view. In this paper, the CAE interface protocol is used to convert the features in the design view into the CAE boundary features [5] in the analysis view. Based on the physical knowledge, an expert system is established to further process those features and generate a robust simulation model with the help of fluid physics features and dynamic physics features in the analysis view.
Download paper
Authors: Lange, Carlos & Ma, Yongsheng.
CFD (Computational Fluid Dynamics) requires strong expertise and extensive training to obtain accurate results. To improve the usability in the complex product development process, two new types of engineering features, fluid physics feature and dynamic physics feature, which convey the simulation intent, are proposed in this paper to achieve CFD solver setup automation and robust simulation model generation in an ideal CAD/CAE integration system. Further, the association between simulation intent and design intent is integrated with another newly defined fluid functional feature in order to achieve the consistency. Consequently, an optimal design could be achieved by considering production operation, manufacturability and cost analysis concurrently. A case study of steam assisted gravity drainage (SAGD) outflow control device (OCD) is presented to show the prospective benefits of the method.
Download paper