Oil & Gas - Maximising Recovery Programme (formerly SHARP) IOR Views

DTI JIP Papers

Outcrop Characterization of Transgressive Sandstone Reservoirs
Presented by Gary Hampson (Imperial College)

This is a Joint Industry Project (JIP) that runs for two years (Feb-06 to Jan-08) and is supported by Chevron, Shell, BP, Energy and Geoscience Institute and DTI. 

The project aims to develop a generic conceptual model and quantitative datasets to characterize all transgressive sandstones by conducting a comparative study of outcrop reservoir analogues to describe facies architecture, internal heterogeneity and rock quality distribution using different shoreline trajectories.  

The presentation started off by discussing the concept of shoreline trajectory and how it can be used to understand transgressive reservoir architecture.  The outcrop analogue datasets were than presented and a subsurface study of these demonstrated the application of the shoreline trajectory concept.  The presenter also highlighted the project’s applications to the Jurassic transgressive sandstone reservoirs of the UK North Sea, like mature brownfield reservoirs in the Northern North Sea (Beryl formation and Brent Group reservoirs) and low permeability reservoirs in Central North Sea (Fulmar formation and Humber group reservoirs).  

During the Q&A session Malcolm Pye from DTI commented on the step up geometry being dependant on bedding thickness. Presenter answered by saying that this was controlled by degree of erosion by overlaying surfaces.

Delegates Networking at DEVEX 2006

Gas Condensate Recovery Studies: Well Deliverability Prediction and Enhancement
Presented by M. Jamiolahmady (Heriot Watt University)

This is a Joint Industry Project (JIP) supported by DTI, BP Columbia, Gaz de France, Norsk Hydro, Statoil and Total UK.

A generalized relative permeability correlation was developed based on relative permeability ratios expressing the combined effects of coupling and inertia.  The reliability of this correlation was calibrated against experimental results. 

A mathematical model for gas condensate flow in perforated regions was also developed and used to generate generalized mechanical and two phase skin factor correlations.  Mathematical models to devise correlations for hydraulic fracturing design optimization were also developed.  The objective of these models was to simulate the perforated/fractured well performance without going into explicit fine grid reservoir simulation.

Uncertainty Quantification & Geological Parameterization
Presented by M. Christie (Heriot Watt University)

This is a Joint Industry Project (JIP) in its second phase supported by BP, BG, Chevron, ConocoPhillips, DTI, ENI, JOGMEC, Shell, TransformSW.

Mike Christie presented some of the techniques developed by this research for quantifying uncertainty.  This was done with the help of two real field cases and another synthetic one.  A computational framework was used to achieve multiple good history matched models for the case of Teal South in the Gulf of Mexico.  Another un-named field model study used diversity of history matched models to show that there was a measurable difference in uncertainty in forecasts of water cut.  The synthetic model study showed that uncertainty forecast can result in over-confidence in the absence of a good spread of history matched models.  

The next phase of the project aims to study geological parameterization by constraining the models to a range of physically allowed reservoir descriptions.

Multi-Rate Transfer Dual Porosity Modelling of Gravity Drainage and Imbibition
Presented by Martin Blunt (Imperial College)

This is a Joint Industry Project (JIP) supported by DTI, Shell and ITF Sponsors.

Martin Blunt highlighted the results of this research that models displacement processes in fractured reservoirs.  The methodology followed for this study consisted of extending the streamline based formulations to incorporate capillary and gravity driven displacement using transfer functions.  According to the study, traditional shape factor geometry-based approach may be inaccurate as it does not explicitly capture the correct average behaviour.  It was suggested that the physics of the process should also be taken into account.  The recovery predictions were highly dependant on the rate constant chosen and therefore suggested that core-scale measurements of the displacement processes should be performed to reduce uncertainties in recovery estimates. 

A multi-rate model accounted for the heterogeneity in the fracture network and allowed for the transfer between fracture and matrix over different time scales.  It also accommodated for variation in fracture spacing and matrix permeability.    The approach was applied to a fractured Chinese oil field (Liu7) to determine the optimal development strategy.

During the Q&A session, the presenter was asked if these formulations are being incorporated in any of the commercial simulators.  Martin answered by saying that work is in progress on making the formulations more refined and sophisticated and  collaboration with Roxar is in process for this.  To another question about working one’s way around this methodology by using an array of shape factors, Martin replied that this would not be able to re-produce the early time behaviour very well.  Replying to a question about the drive mechanism in the Chinese oil field, Martin said that over the years the operator has changed its plans, but currently it is primary production with small water injection and no aquifer support.  The remote location of the field limits any gas re-injection.

Analyzing Contaminated Reservoir Fluid Samples and Modelling Asphaltine Precipitation
Presented by F. Gozalpour (Heriot Watt University)

This is a Joint Industry Project (JIP) sponsored by DTI, Gaz de France, Marathon, Schlumberger, Shell and Total.

This presentation discussed the findings of the current phase (started in Nov-2002) of this research project where the focus has been on reservoir fluid sampling, measurement and modelling of gas tracer partitioning, HPHT fluids and asphaltine precipitation prediction in gas condensate reservoirs.  

The use of tracer partitioning between liquid and vapour in oil-based mud allowed retrieving original reservoir fluid composition, and consequently original reservoir fluid properties, when significant mass transfer occurred between the reservoir fluid and mud filtrate.  Optimum probe location (at the top of the layer beneath the non-permeable boundary) and desired formation geology (low vertical to horizontal permeability) were also determined.

Furthermore, an EOS and solubility based thermodynamic model for predicting asphaltine precipitation was also developed which uses an excess Gibbs energy based mixing rule.  The model showed promising results against measured data.

Water Alternating Gas Injection Processes’
Presented by M. Sohrabi (Heriot Watt University)

This is a Joint Industry Project (JIP) supported by DTI, BHP, BP, Norsk Hydro, Shell, Statoil and Total.

This research project aims to study the Water Alternating Gas (WAG) injection flow processes and predict accurately the associated three phase flow performance and recovery.  This presentation discusses the results of the current phase of this project that concentrated on core flood studies, numerical simulation and pore-scale network modelling. 

Relative permeability curves for both drainage and imbibition were measured for immiscible and near miscible conditions.  The experiments were numerically simulated to account for the effects of capillary pressure and heterogeneity.

The parameters for the pore scale network model were determined based on single and two phase measurements from a North Sea reservoir.  Three phase gas floods were than simulated starting at different initial water saturations.  Three phase relative permeabilities and capillary pressures were thus calculated along the resulting saturation paths.  Modelling of oil wetting films around gas in oil-wet pores was found to have a large bearing on the resulting three phase saturation paths.  Some example phase paths for WAG processes were shown but a detailed account of these WAG calculations, effects of contact angles, hysteresis are planned to be presented at a later stage.

During the Q&A session, Colin Cranfield from DTI asked about the relevance of the pore throat size for these models with reality.  Sohrabi accepted the validity of the comment but said that it is important in such experiments to have a magnification that allows good visualization and thus the pore throat sizes have to be larger in such models compared to actual reservoir geometries.

Gas Hydrates and Flow Assurance
Presented by B. Tohidi (Heriot Watt University)

This work is funded by Clariant Oil Services, Petrobras, Shell, Statoil, Total and the UK
Department of Trade and Industry

The Centre for Gas Hydrate Research at the Heriot-Watt University has actively pursued gas hydrate research since 1986. The current phase of Gas Hydrate and Flow Assurance JIP, started on 1 December 2005, is an integrated experimental and modelling study addressing several important topics in avoiding Flow Assurance problems in oil and gas production and transportation.  These include Gas Hydrates in Low Water Content Gases, Transportability of Water/Oil Mixtures, Hydrate Equilibrium Measurements Including Oil Systems, Inhibitor Distribution in Water and Hydrocarbon Phases, Wax and Wax Inhibitors, Gas Hydrates in Water-Flooded Oil Reservoirs and Database on Hydrates and Phase Equilibria.

This paper presents the results of recent research on two of the these topics (i.e., Gas
Hydrates in Low Water Content Gases and Wax and Wax Inhibitors) and discusses its achievements in obtaining useful experimental data, important physical parameters, improved correlations for inhibition predictions and information on potential influence of wax on hydrates and vice versa.

Local Methods for Quality Enhancement of Pre-Stack Seismic Data
Reserve Mirko van der Baan (University of Leeds)

This project ‘Blind Identification of Seismic Signals’ is sponsored by the BG group, BP, Chevron, the Department of Trade and Industry and Shell.

A critical issue in seismic data processing is the enhancement of signals embedded in background noise, particularly of pre-stack gathers. Certain pre-stack techniques provide some noise reduction, but often result in an increase of the noise level at high frequencies. This research focuses on the use of coherency-based techniques to improve the quality of the seismic image.

Singular Value Decomposition (SVD), a coherency-based technique that provides both signal retrieval and noise suppression, has been used in this research to improve the signal-to-noise ratio of pre-stack seismic gathers by applying it locally to cope with signals that vary both with time and offset. When combined with Independent Component Analysis (ICA), it performs better than f -x de-convolution in removing background noise but less well in enhancing the lateral coherency of weak events and/or events with conflicting dips. Combining f -x de-convolution with SVD/ICA signal enhancement overcomes the main weaknesses associated with each individual method and leads to the best results.