Reservoir Fluid (PVT) Analysis - Value to Appraisal / Field - - PowerPoint PPT Presentation

t: +44 (0) 7771 881182 e: info@petrophase.com

www.petrophase.com

Reservoir Fluid (PVT) Analysis - Value to Appraisal / Field Development Planning Brian Moffatt

PVT Information Key for all areas of Field Development Exploration

• Composition for economics

Appraisal

• Contaminants
• Flow Assurance

Development

• Phase Behaviour for

Reservoir Simulation

Production

• Composition monitoring

PVT Information

How to Maximise the Value of PVT Information?

Issues from Linkedin PVT Forum Questions Value of PVT

Introduction - PVT Concerns

2 4 6 8 10 12 14 16 18 20

Understanding PVT Data EOS Modelling Methods PVT and Reservoir Behaviour Equipment Sampling QC Methods Training

Questions

Forum replies focus on:

• Data QC Methods
• Sampling

Introduction – PVT Concerns

2 4 6 8 10 12 14 16 18 20

Understanding PVT Data EOS Modelling Methods PVT and Reservoir Behaviour Equipment Sampling QC Methods Training

Questions Replies/Question

How to Maximise the Value of PVT Information?

• PVT Data QC
• Uncertainties from Sampling
• Storage Issues
• Uncertainties from PVT Lab Measurements
• Understand the Data in Context
• Modelling Key Information
• Focus on Matching Key Data
• Correct handling MWs
• Poor EOS performance for oil compressibility and viscosity
• Mapping reservoir simulation results to a surface model
• Which PVT uncertainties can most affect Development?

This Presentation

PVT Data QC Maximising the Value of PVT Data

PVT Data QC

PVT DATA QC Traditional QC Sampling Conditions Well Characteristics Field GOR vs Lab data Sample Quality Air /OBM Contamination Opening Pressures of Samples Sample Compositions Equilibrium Plots Data trends Lab Measurements Consistency Material Balance Equilibrium Plots Context / Application Agreement with Field Data

Sampling

• Bottomhole-two phase flow into sampler
• Formation tester-OBM Contamination
• Separator-Reservoir two phase flow,

Recombination GOR, Liquid Carryover Storage

• Contaminant absorbtion

Measurement Errors

• Sample handling-loss of heavy ends from gas

samples

Where do PVT Data Errors Arise ?

QC for Sampling Errors

Maximising the Value of PVT Information

Problem Areas

• Bottomhole-two

phase flow into sampler

• Commingled flow

from different intervals

QC: Bottomhole Flowing Samples

Information Obtained

• Formation pressure and

pressure gradient (fluid type)

• Estimate formation

permeability.

• Sample compositions

Possible Problems

• Two phase flow from poor

probe contact

• OMB contamination

QC: Formation Tester

QC FT: OBM Contamination

• GC trend analysis: hump in the compositional analysis, especially observed in

the carbon number range of the oil based mud components (C15-C20).

QC FT: Poor Compositions

9950 10000 10050 10100 10150 10200 10250 10300 10350 10400

5550 5600 5650 5700 TVD SS ft Pressure (psia)

Data PVT Report Oil

Use sample composition in an EOS Analysis to compare predicted and measured values for

• Surface GOR
• Phase Behaviour

Compare PVT Lab Densities with Densities from Pressure Gradients

QC: Surface Sampling

0.0 2.0 4.0 6.0 8.0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 30 C 50 C

Separator GOR is highly dependent on surface conditions. Should not affect recombined fluid.

QC: Surface Sampling

Pressure, Bar bbl/MMscf One lean condensate at different conditions

1,000 2,000 3,000 4,000 5,000 6,000 7,000 500 1000 1500 2000 GOR Mscf/bbl

WHP psia CGR vs Sep Press

What causes CGR scatter? Conditions? Wellstream?

QC: Surface Sampling

CO2 N2 C1 C2 C3 iC4 nC4 iC5 nC5 C6 Benz C7 Tol C8

0.5 1 1.5 2 2.5 3 3.5 4 4.5

• 2

2 4 6 Log10 (K*P) Temp Function

Hoffmann-Hocott Equilibrium Plot

Data Theory

QC: Surface Sampling

Equilibrium Plot Between Surface Liquid and Gas Compositions. Identifies

• Liquid Carry-over
• Sample Handling

Loss of heavy ends

• Poor Temperature/

Pressure Readings

Trend for Carryover Trend for Heavy end Losses

Slide 19

What if after QC of Surface and BH samples, there are no

• bvious errors but the Compositions Disagree AGAIN!

QC Data in Context: Strange GC

0.001 0.01 0.1 1 10 100 % MOL

BHS1 BHS2 BHS3 Separator

Slide 20

• Initial GOR was steady at

around 8,000 scf/bbl and samples gave a typical Gas Condensate behaviour

• However recombined

Separator Sample gives Psat> Pres

• This was a low

Permeability Formation with high drawdown

1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 10,000

1 2 3 4 5 6 7 8 9 10

GOR scf/bbl Flow Period

Pres

QC Data in Context: Strange GC

Slide 21

• At lower rates and lower

drawdowns the tested GOR reduced

• The API and Liquid

colour suggested the fluid maybe a Volatile oil

• An EOS analysis giving a

fluid with Psat= Pres gas a Volatile Oil with GOR value of 2000 scf/bbl

• FLUID IS VOLATILE OIL!

2,000 4,000 6,000 8,000 10,000

2 4 6 8

GOR scf/bbl

Gas Rate mmscf/d

QC Data in Context: Strange GC

QC for Sample Storage and PVT Measurements

Maximise the Value of PVT Information

Pressures of Sample Bottles drop during storage due to cooling

• Where groups of samples available the highest pressure sample is

less likely to have suffered leakage and compositional changes

• With pressure drop can get deposition of asphaltenes/ sometimes

reversible Contaminant absorbtion a problem in non conditioned bottles

QC:Sample Storage

Consistency Checks used for Common Lab Measurements

• CVD/DL Material Balance
• EOS Modelling for reality checks

Consistency Checks routinely carried out by PVT labs, data quality now generally excellent. However historical data and data from unknown labs can still have errors.

QC:PVT Lab Measurements

AT P=0, Z-factor approaches Unity

Modelling Key Information

Maximising the Value of PVT Information

Which data do you match to? The “Best Fit” may not match well in the are of interest, e.g. if the reservoir does not drop below the saturation pressure

Modelling Key Information

PVT labs measure volumetrics well, however EOS can struggle with compressibilities. EOS models are particularly limited in modelling near critical fluids. Unrealistic phase envelopes can arise. Beware of using different compositions in a well matched EOS!

Modelling Key Information

0.660 0.670 0.680 0.690 0.700 0.710 0.720 0.730 0.740 0.750 5000 6000 7000 8000

Density g/cc

Pressure (psia) DATA SRK

Unlikely Critical Behaviour

Conversion difficulties in transferring from reservoir modelling software to processing modelling software! Reservoir Engineer's Process Engineer’s perspective perspective

Modelling Key Information

Matching viscosity using the LBC correlation is highly dependent on densities. Poor densities gives poor viscosities away from control points, and also for the gas !

PVT Modelling Errors -Viscosities

1000 2000 3000 4000 5000 6000 7000 8000 9000 200 400 600 800

Viscosity cp

Rs scf/bbl

B&R Kartoatmodjo Kartoatmodjo HO P&F

Liquid viscosities are not well predicted by EOS and so often correlations are

• used. For heavy oil the

errors can be >100%.

PVT Modelling Errors - MW

• Samples are prepared

Gravimetrically

• Response of GC Detectors are

Proportional to Mass

– Internal standards are added by weight

30

Increasing MW Oil

PVT Modelling Errors - MW

SCN31

• Average Molecular Weight for a Fraction not Known
• Each Fraction has Complex Mix of Compounds
• Different Service Companies may use Different Sets

100 120 140 160 180 200 220 240 260 280 C9 C10 C11 C12 C13 C14 C15 C16 C17 C18 C19

Fraction MW

Core Labs Petrobras Expro

• Volumetrics for economics
• Measured GORs
• Phase Behaviour for Reservoir Simulation
• Contaminants, Flow Assurance Issues
• Viscosities
• Compositions

Which PVT uncertainties data can most affect Development?

PVT labs measure reservoir condition densities to better than 1%. Insensitive to compositional errors from

• sampling. Errors small

compared to GRV and Sw errors. However, surface liquid volumes and hence STOIIP strongly influenced by Separator Conditions.

Volumetrics

0.0 2.0 4.0 6.0 8.0 10 20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180 30 C 50 C

CGR vs Sep Press

The GOR is often chosen for modelling from a single recombined sample! Is this sample consistent with the rest of the test data? Often ignore much relevant test data.

GOR

1,000 2,000 3,000 4,000 5,000 6,000 7,000 500 1000 1500 2000 GOR Mscf/bbl

WHP psia

PVT labs measure volumetrics well, however still a need to QC particularly old data Sampling errors can lead to unrepresentative phase behaviour.

Phase Behaviour for Reservoir Simulation

Pres

PVT labs measure/calculate gas viscosities to +/- 1%. Liquid viscosities to +/- 5%. Unlikely to be important even in tight formations as permeability errors are larger. Matching viscosities using the LBC correlation gives values are highly dependent on densities, poor densities gives poor viscosities away from control points, and also for the gas !

Viscosities

1000 2000 3000 4000 5000 6000 7000 8000 9000 200 400 600 800

Viscosity cp

Rs scf/bbl

B&R Kartoatmodjo Kartoatmodjo HO P&F

Liquid viscosities are not well predicted by EOS and so often correlations are

• used. For heavy oil the

errors can be >100%, beware!

Contaminants and flow assurance issues can lead to costly topsides processing facilities and can constrain export options. Huge cost implications, cf Buzzard.

Contaminants and Flow Assurance

H2S against Cumulative Gas Production 5 10 15 20 25 30 35 40 45 50 0.0 1.0 2.0 3.0 4.0 5.0 6.0 7.0

Cumulative gas (MMSCF) H2S (PPM)

A Surprise!

Contaminants and Flow Assurance

Conditions Classification by Habitat Compounds Possible Biogenic compounds formed < 70 oC Me3As, Hg (element) and Me2Hg, MeSH, Me2S, Maturation products formed < ~140 oC CO2, H2O, H2S, R-SH, R-S-R’, R-S-S-R’ Thiophenes, tetrahydrothiophenes, benzothiophenes (R and R’ are alkyl groups, methyl, ethyl propyl etc) Thermally stable products > ~140 oC S(vap), Hg, CO2, H2S, COS, N2, H2O (as steam or liquid)

Deeper, hotter & high pressure

Conclusions

Main Problems!

• GOR measurements
• Phase behaviour from poor samples
• Poor modelling of heavy oil viscosities and compressibilities

Be aware!

• Contaminants
• Wax, scale and asphaltene deposition
• Compositions to indicate compartmentalisation
• Small heavy end compositions

Placing the PVT data in context is one of the best methods of Data QC

Capture the Real Picture!

http://www.petrophase.com

info@petrophase.com

The End - Thank you!