An Interactive 5-Day Training Course
Upscaling from Geological Model to Simulation Grid
Upscaling Models Without Losing Flow Behaviour
Secure your spot now!
Duration
5 Days
Price
$ 3,500
Starting From
04 Jan 2027
Home Geology and Geoscience Upscaling from Geological Model to Simulation Grid
Upscaling from Geological Model to Simulation Grid
Have Questions ?
Upscaling from Geological Model to Simulation Grid - GG-UPSC-PEA27
| Code | Date | Time | Duration | Location | Currency | Early Bird Fee Per Person |
|---|---|---|---|---|---|---|
| GG-UPSC-PEA27 | 04 - 08 Jan 2027 | 10 AM CST | 4 Hours Per Day |
Online |
USD |
3500 |
Need this for a group? We deliver the same course in-house — face-to-face at your location or online — tailored to your assets and team level. Contact info@peassociations.com
Boost your team's skills and your budget! Enjoy group discounts for collaborative learning. Send an inquiry to info@peassociations.com.
Upscaling from Geological Model to Simulation Grid
This training covers upscaling from fine geological models to simulation grids. It works through upscaling objectives and what must be preserved, coarse grid design, static property averaging, permeability upscaling methods, flow based and steady state methods, upscaling of relative permeability and capillary pressure, fault and barrier representation, validation of the upscaled model and the diagnosis of upscaling failures.
Description
A geological model may contain fifty million cells and a simulation model two million. Reducing one to the other without changing the flow behaviour is the upscaling problem, and it is not solved by averaging. Permeability does not average arithmetically in the general case, thin barriers disappear when averaged into thick cells, and the heterogeneity that causes bypassing in the fine model may be smoothed away entirely in the coarse one. The result is a simulation model that flows differently from the geology it was built from.
This training treats upscaling as a technical problem with defined answers. Objectives are established first: what the upscaled model must preserve, which depends on what the simulation is for. Coarse grid design follows, covering layer grouping, areal coarsening, where to keep resolution and where it can be lost. Static property averaging is then developed, covering porosity, net to gross and saturation, which are straightforward, before permeability, which is not. Permeability upscaling is then addressed in detail, covering arithmetic, harmonic and geometric averaging and their validity, power averaging, renormalisation and tensor methods. Flow based upscaling follows, covering single phase and steady state methods and their boundary condition dependence. Relative permeability and capillary pressure upscaling are then developed, including pseudo-function generation. Fault and barrier representation follows. The training closes with validation and failure diagnosis.
The averaging method depends on the flow geometry, not on preference. Permeability averages arithmetically for flow parallel to layering, harmonically for flow perpendicular to it, and geometrically for randomly heterogeneous media. A single layered cell therefore needs arithmetic averaging for horizontal permeability and harmonic for vertical, and using one method for both is simply wrong. The errors are large and systematic: harmonic averaging of horizontal permeability can understate it by orders of magnitude where a high permeability streak is present.
Thin shales matter out of all proportion to their thickness and upscaling destroys them. A continuous half-metre shale within a twenty-metre sand creates a vertical permeability barrier that controls coning, injection conformance and gravity drainage. Averaged into a five-metre coarse cell, it reduces vertical permeability slightly and removes the barrier. The field behaves as though the barrier is there and the model does not, which is a history match problem with no parameter solution because the geometry is wrong. Barriers have to be preserved deliberately, through layer boundaries placed at them or through explicit transmissibility multipliers.
Flow based upscaling is more accurate and depends on its boundary conditions. Running a fine scale flow simulation through each coarse cell gives an effective permeability that reproduces the fine scale flux, which is a better answer than any averaging formula. But the result depends on the boundary conditions imposed, and the pressure, periodic and no-flow options give different answers, sometimes considerably different for strongly heterogeneous cells. Choosing boundary conditions appropriate to the expected flow direction, and reporting the choice, is part of the method.
Finally, validation is a fine against coarse flow comparison and is often skipped. The only reliable check on an upscaled model is to run a flow simulation on the fine model and on the coarse one under the same conditions and compare the production profiles, breakthrough times and recovery. Where they differ, the upscaling has lost something. This is computationally expensive for the full field, but can be done on sector models, and a team that has never made this comparison does not know how much its upscaling changes the answer.
By the end of this training, participants will be able to:
- Define what an upscaled model must preserve for a given simulation purpose
- Design coarse grids that retain flow-critical resolution
- Average static properties correctly including net to gross and saturation
- Select permeability averaging methods appropriate to the flow geometry
- Apply power averaging, renormalisation and tensor methods
- Apply flow based and steady state upscaling and select boundary conditions
- Generate pseudo relative permeability and capillary pressure functions
- Preserve thin barriers and baffles through upscaling
- Represent faults and transmissibility in the coarse grid
- Validate upscaled models by fine against coarse flow comparison
The training is upscaling practice on fine scale models with known flow behaviour. Participants upscale the same model by different methods and compare the resulting flow against the fine scale reference. The effect of removing thin barriers is quantified directly. Boundary condition choices in flow based upscaling are varied and the spread in effective permeability measured. Deliberately poorly upscaled models are supplied for diagnosis.
Organisations sending participants to this training will:
- Produce simulation models that flow like their geological models
- Preserve barriers and baffles that control field behaviour
- Select averaging methods on technical grounds
- Reduce history matching effort caused by upscaling errors
- Validate upscaling rather than assuming it
- Improve forecast reliability from simulation
Participants will:
- Upscale permeability correctly for the flow geometry
- Design coarse grids that keep what matters
- Generate and apply pseudo functions
- Preserve thin barriers deliberately
- Validate upscaled models properly
- Diagnose upscaling related history match problems
- Reservoir modellers and geomodellers
- Reservoir engineers building simulation models
- Technical staff handing models between static and dynamic teams
- Petrophysicists working on property models
- Simulation engineers diagnosing history match problems
- Technical staff reviewing simulation models
- Geoscientists responsible for static to dynamic handover
Module 1 - Upscaling Objectives
- Why upscaling is necessary
- What the coarse model must reproduce
- Purpose dependence of upscaling requirements
- Volume preservation requirements
- Flow behaviour preservation requirements
- Breakthrough time and sweep
- Recovery factor preservation
- Acceptable error levels
- Upscaling as information loss
- Deciding what can be lost
- Setting upscaling specifications
Module 2 - Coarse Grid Design
- Coarsening ratio selection
- Areal coarsening and its effects
- Vertical layer grouping
- Grouping by flow unit
- Layer boundaries at barriers
- Preserving high permeability streaks
- Local refinement around wells
- Refinement in high flow regions
- Grid alignment with flow direction
- Computational size constraints
- Testing alternative coarse grids
- Documenting grid design decisions
Module 3 - Static Property Averaging
- Porosity volume weighted averaging
- Net to gross averaging
- Combining porosity and net to gross
- Saturation averaging and pore volume weighting
- Facies upscaling and dominant facies methods
- Facies proportion preservation
- Rock type upscaling problems
- Mixing rock types in one coarse cell
- Volume preservation checking
- Pore volume and hydrocarbon volume checks
- Static quality control of the coarse model
Module 4 - Permeability Averaging Methods
- Why permeability does not average arithmetically
- Arithmetic averaging and parallel flow
- Harmonic averaging and series flow
- Geometric averaging and random media
- Bounds on effective permeability
- Power law averaging and exponent selection
- Directional averaging for layered cells
- Renormalisation methods
- Permeability tensors and off-diagonal terms
- Anisotropy introduced by upscaling
- Vertical permeability upscaling
- Method selection by heterogeneity type
Module 5 - Flow Based Upscaling
- Flow based upscaling principle
- Single phase flow simulation per cell
- Pressure boundary conditions
- Periodic boundary conditions
- No-flow boundary conditions
- Sensitivity of results to boundary conditions
- Local against global upscaling
- Extended local methods
- Steady state upscaling
- Directional flow simulations
- Computational cost management
- When flow based upscaling is justified
Module 6 - Dynamic Function Upscaling
- Why relative permeability needs upscaling
- Pseudo relative permeability concept
- Dynamic pseudo generation methods
- Kyte and Berry method
- Stone and other pseudo methods
- Capillary pressure upscaling
- Capillary and viscous dominated regimes
- Capillary equilibrium assumptions
- Transition zone representation in coarse cells
- Pseudo function validity limits
- Rate dependence of pseudo functions
- Applying pseudo functions in simulation
Module 7 - Barriers, Baffles and Faults
- Thin shale significance for flow
- Barrier loss through averaging
- Preserving barriers at layer boundaries
- Transmissibility multipliers for barriers
- Baffle representation
- Fault transmissibility in the coarse grid
- Juxtaposition after coarsening
- Fault seal representation
- Stair step against pillar fault representation
- Sub-seismic fault effects
- Fracture upscaling in outline
- Checking barrier representation
Module 8 - Validation and Diagnosis
- Fine against coarse flow comparison
- Sector model validation
- Production profile comparison
- Breakthrough time comparison
- Recovery factor comparison
- Acceptance criteria for upscaling
- Diagnosing excessive numerical dispersion
- Diagnosing lost barrier effects
- Diagnosing permeability averaging errors
- Grid orientation effects after coarsening
- History match problems caused by upscaling
- Documenting the upscaling and its validation
Upon successful completion of this training course, delegates will be awarded an official Certificate of Completion issued by the Petroleum Engineers Association (PEA), an ISO 9001:2015 certified training organization. The certificate carries 10 Credits and formally records the total learning hours completed.
Each certificate is signed by the Course Facilitator and the CEO of the Petroleum Engineers Association, and serves as verifiable proof of professional training that delegates can present to employers and professional bodies worldwide.
Frequently Asked Questions
All course bookings made through PEA are strictly non-refundable. By registering for a course, you acknowledge and accept that all fees are payable in full and are not subject to refund under any circumstances, including changes in personal or professional commitments or partial attendance.
PEA reserves the right to make reasonable adjustments to course content, trainers, or schedules where necessary, without entitling delegates to a refund. Comprehensive details of each course — including objectives, target audience, and content — are clearly outlined before enrolment, and it is the responsibility of the delegate to ensure the course's suitability prior to booking.
For any inquiries related to cancellations or bookings, please contact our support team, who will be happy to assist you.
