Slide 1
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Slide 1
SS TSS-02Significant Factors in SelectingThermodynamic Methods for aSimulation Model
John CunninghamNA Tech [email protected]
@InvensysOpsMgmt / #SoftwareRevolution
/InvensysVideos
social.invensys.com
© 2013 Invensys. All Rights Reserved. The names, logos, and taglines identifying the products and services of Invensys are proprietary marks of Invensys or its subsidiaries.All third party trademarks and service marks are the proprietary marks of their respective owners.
John CunninghamNA Tech [email protected]
/InvensysOpsMgmt
/Group/PRO/II
Introduction
With PRO/II version 3.3, SimSci introduced a DOS ThermodynamicExpert System. It was not carried over to the later Windowsversions for a number of reasons.
We have regularly given “Thermodynamic Selection” presentationssince that time.
Recently, researchers at IFP have published a similarThermodynamic Selection approach in book form; this is anadaptation of that approach.
Slide 3
With PRO/II version 3.3, SimSci introduced a DOS ThermodynamicExpert System. It was not carried over to the later Windowsversions for a number of reasons.
We have regularly given “Thermodynamic Selection” presentationssince that time.
Recently, researchers at IFP have published a similarThermodynamic Selection approach in book form; this is anadaptation of that approach.
Introduction
”SELECT THERMODYNAMIC MODELS FOR PROCESS SIMULATION”:A practical Guide using a Three Steps Methodology,” Jean-Charles deHemptinne, Jean-Marie Ledanois, Pascal Mougin, Alain Barreau, EditionTechnip, Paris 2012.
http://books.ifpenergiesnouvelles.fr/ebooks/thermodynamics/table_of_contents.html
I will be showing figures from both the book and PRO/II
Slide 4
”SELECT THERMODYNAMIC MODELS FOR PROCESS SIMULATION”:A practical Guide using a Three Steps Methodology,” Jean-Charles deHemptinne, Jean-Marie Ledanois, Pascal Mougin, Alain Barreau, EditionTechnip, Paris 2012.
http://books.ifpenergiesnouvelles.fr/ebooks/thermodynamics/table_of_contents.html
I will be showing figures from both the book and PRO/II
Today’s Presentation
Introduction
From Fundamentals to Properties
From Components to Models
From Phases to Method(Model) Selection
Slide 5
Introduction
From Fundamentals to Properties
From Components to Models
From Phases to Method(Model) Selection
Property Interconnectivity
Slide 6
A Three Step MethodologyThermodynamics Universal Equations
Properties
Slide 7
Methods – Models - Data
Components Phases
Choice of Models in Each Phase
Methods; The Physical Model
InputParameters
OutputParameters
ThermodynamicMethod
Model 1
Slide 8
Model 2
Algorithm
= Vapor + EntrainedLiquid
Any phase can beentrained to any
other phase
Introduction: What is “the Model”
Slide 9
= Vapor + EntrainedLiquid
Any phase can beentrained to any
other phase
After entrainment, products may not be at equilibrium
Important Questions
What properties do we have? What more do we need?
What are the mixture components?
Where are the process conditions located with respect to thephase envelope?
Slide 10
What properties do we have? What more do we need?
What are the mixture components?
Where are the process conditions located with respect to thephase envelope?
From Fundamentals to Properties
An unambiguous definition of a thermodynamic state may be given by:
Temperature
Pressure
Composition
Physical State (ideal gas, vapor, liquid, solid)
Method selection decisions can be made on the basis of the abovebased on both the properties available and those required.
Slide 11
An unambiguous definition of a thermodynamic state may be given by:
Temperature
Pressure
Composition
Physical State (ideal gas, vapor, liquid, solid)
Method selection decisions can be made on the basis of the abovebased on both the properties available and those required.
From Fundamentals to Properties
The Gibbs phase rule gives insight to the number of phases that weneed to take into account:
F = C − φ + 2 −R
F= degrees of freedom
C=number of components
Φ=number of phases
R=restraints (azetrope, critical, etc.)
For a single component:
F = 3 – φ; so 2 phase regions are lines, criticals, 3 phase regions,points
Slide 12
The Gibbs phase rule gives insight to the number of phases that weneed to take into account:
F = C − φ + 2 −R
F= degrees of freedom
C=number of components
Φ=number of phases
R=restraints (azetrope, critical, etc.)
For a single component:
F = 3 – φ; so 2 phase regions are lines, criticals, 3 phase regions,points
From Fundamentals to Properties
Slide 13
From Fundamentals to Properties
• Reliable Pure Component Properties can be obtained fromthe Simulators Library, most of the data in PRO/IIcomponent libraries originate from either DIPPR orPPDS/NEL in the UK
• Many non-library component properties can be estimatedvia group contribution methods available in PRO/II or TDM
• There is a need to cross check Pure component values
Slide 14
• Reliable Pure Component Properties can be obtained fromthe Simulators Library, most of the data in PRO/IIcomponent libraries originate from either DIPPR orPPDS/NEL in the UK
• Many non-library component properties can be estimatedvia group contribution methods available in PRO/II or TDM
• There is a need to cross check Pure component values
Mixtures with limited numbers of components
Slide 15
Mixtures with limited numbers of components
50
60
70
80
90
100
Tem
per
atu
re,
C
T-X-Y Plot for PROPANE and PENTANE
Slide 16
0
10
20
30
40
50
0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
Tem
per
atu
re,
C
Composition, Mole Fraction PROPANE, (P = 0.50000 MPa)
Mixtures with limited numbers of components
Slide 17
Mixtures with limited numbers of components
Slide 18
Lets talk about compositionClassifications:
SIMPLE: Light ends, large experimental database “REFPROP types”
LIGHT: Hydrocarbons to C10, other well known industrial chemicals
HEAVY: Hydrocarbons >C10, little or sparse experimental data
PSEUDO: Derived from petroleum assays, characterization methodstypically known
Different levels of verification are necessary for verification of propertyvalidity
Slide 19
Classifications:
SIMPLE: Light ends, large experimental database “REFPROP types”
LIGHT: Hydrocarbons to C10, other well known industrial chemicals
HEAVY: Hydrocarbons >C10, little or sparse experimental data
PSEUDO: Derived from petroleum assays, characterization methodstypically known
Different levels of verification are necessary for verification of propertyvalidity
Lets talk about composition
Methane, a “simple” component
2-methyl benzothiopheneA “heavy” component
Slide 20
API needed better Vp datafor low sulfur Diesel
Lets talk about composition
Next Question:
What is the key composition range?
Most cases it’s the majority components that impact thecalculation
On occasion it can be an impurity that matters
Slide 21
Next Question:
What is the key composition range?
Most cases it’s the majority components that impact thecalculation
On occasion it can be an impurity that matters
Lets talk about compositionNext Questions:
Is the mixture ideal or non-ideal, What type and to whatdegree is the extent of this non-ideality
How do we think about non-ideality?
Raoult’s Law:
Activity coefficient:
Slide 22
Next Questions:
Is the mixture ideal or non-ideal, What type and to whatdegree is the extent of this non-ideality
How do we think about non-ideality?
Raoult’s Law:
Activity coefficient:
Lets talk about composition
• When non-ideality is due to attraction or repulsionbetween molecules we call it enthalpic (or residual)effects. Attraction results in positive deviationsfrom ideality while repulsion tends to result innegative deviations
• If the non-ideality is due to molecular size it isknow as entropic (or combinatorial) effects
Slide 23
• When non-ideality is due to attraction or repulsionbetween molecules we call it enthalpic (or residual)effects. Attraction results in positive deviationsfrom ideality while repulsion tends to result innegative deviations
• If the non-ideality is due to molecular size it isknow as entropic (or combinatorial) effects
Lets talk about composition
Slide 24
Lets talk about composition
Slide 25
Lets talk about composition
Slide 26
Finally different Methods may be good fordifferent phases/properties
Slide 27
Considerations in the Critical Region: Simulation Considerations
Slide 28
Other considerations in the CriticalRegion: property considerationsHeat capacity example, Cp going to infinity
Slide 29
Considerations in the Critical Region:Complex Phase BoundariesPhase boundary for water/ethyl benzene: representative of 20 or so
more complex phase boundary systems
Slide 30
Considerations in the Critical Region:Thermodynamic Model Considerations
Liquid Activity models (or Gamma Phi approach) uses pure componentvapor pressures to calculate phase equilibria
Vapor Pressure is undefined above the critical point
Henrys Law is used
Replaced by:
K
y
x
PvRT
P P
Pii
i
il
isat
isat i
l
isat
iv
exp
Slide 31
Liquid Activity models (or Gamma Phi approach) uses pure componentvapor pressures to calculate phase equilibria
Vapor Pressure is undefined above the critical point
Henrys Law is used
Replaced by:
K
y
x
PvRT
P P
Pii
i
il
isat
isat i
l
isat
iv
exp
Ky
x
H
Pii
i
i mix ,
We will modify the Decision tree below to be more PRO/II
centric
Slide 32
Working through the HC Portion of the Tree
Pretty Easy
The Yes side of only HC?
i.e. H2/nC20 or Methane/Tetralin
Our First Solution: I like CEOSwhere fitted alphas and Kijs areavailable
The No side of only HC;
Yes for azeotropy?
Next solution: Fitted CEOS orPPR78 (French Version of PRwith generated Kijs)
Slide 33
Pretty Easy
The Yes side of only HC?
i.e. H2/nC20 or Methane/Tetralin
Our First Solution: I like CEOSwhere fitted alphas and Kijs areavailable
The No side of only HC;
Yes for azeotropy?
Next solution: Fitted CEOS orPPR78 (French Version of PRwith generated Kijs)
Working through the HC Portion of the Tree
Slide 34
Working through the HC Portion of the Tree
No for azeotropy?
Yes for close to critical?
Next solution: Fitted CEOS orPPR78
No for close to critical?
No for P>0.1 MPa (1 atm.)
Next solution: CEOS or PPR78
Yes for P>0.1 MPaYes for pseudo?
BK10; No, CEOS w/fitted alphas
Slide 35
No for azeotropy?
Yes for close to critical?
Next solution: Fitted CEOS orPPR78
No for close to critical?
No for P>0.1 MPa (1 atm.)
Next solution: CEOS or PPR78
Yes for P>0.1 MPaYes for pseudo?
BK10; No, CEOS w/fitted alphas
Working through the non-HC Portion of the Tree
Pretty Easy again
Yes for size asymmetry?
Yes or No for P>1 MPa
Both answers essentially leadto Polymer thermo solutions
No for size asymmetry?
No for Water + HC?
Slide 36
Pretty Easy again
Yes for size asymmetry?
Yes or No for P>1 MPa
Both answers essentially leadto Polymer thermo solutions
No for size asymmetry?
No for Water + HC?
Working through the non-HC Portion of the Tree
No for water + HC?
I would like this better as water +organics
Low pressure use an activitycoefficient approach, high pressures anadvanced CE0S
Parameters are Absolutely Necessary!
Slide 37
No for water + HC?
I would like this better as water +organics
Low pressure use an activitycoefficient approach, high pressures anadvanced CE0S
Parameters are Absolutely Necessary!
Working through the non-HC Portion of the Tree
Yes for water + HC?
I would like this better as water +organics
No for electrolytes?
Advanced CEOS
Yes for electrolytes?
Electrolyte Methods in PRO/II areOLI’s Pitzer based method and thee-UNIQUAC MSE available foradditional fees
Slide 38
Yes for water + HC?
I would like this better as water +organics
No for electrolytes?
Advanced CEOS
Yes for electrolytes?
Electrolyte Methods in PRO/II areOLI’s Pitzer based method and thee-UNIQUAC MSE available foradditional fees
Slide 39
Important Background Information
Know what methods are available and what data are present inthe property databanks
Keep in mind what the intended usage for each method is andwhen you’re “stretching” those limits
Help is available:
• On line documentation:
– Reference Manual, Volume 1
– Keyword input manual; Component and Thermodynamics volumes
– Pay the price! Know what information is available.
• You can call the NA hotline number at any time, you will get anexperienced engineer
Slide 40
Know what methods are available and what data are present inthe property databanks
Keep in mind what the intended usage for each method is andwhen you’re “stretching” those limits
Help is available:
• On line documentation:
– Reference Manual, Volume 1
– Keyword input manual; Component and Thermodynamics volumes
– Pay the price! Know what information is available.
• You can call the NA hotline number at any time, you will get anexperienced engineer
Important Background - ToolsAvailable in PRO/II’s GUI (Provision)• BVLE
–Viewing binary Phase Behavior
• HCURVE–Tracking Pure and mixture properties
• PHASE–Important for light ends processing and production
• Regression via TDM or PRO/II–Developing correlation Coefficients
• Component Properties
Slide 41
Available in PRO/II’s GUI (Provision)• BVLE
–Viewing binary Phase Behavior
• HCURVE–Tracking Pure and mixture properties
• PHASE–Important for light ends processing and production
• Regression via TDM or PRO/II–Developing correlation Coefficients
• Component Properties
Important Background - Tools
TDM• Database Manipulation
– Creation of library files
– Loading the various pure component and mixture databases
• Property Regression
– Converting data into correlated forms for databank storage for use insimulation
• VLE Regression
–Available in TDM 3.0; PRO/II 8.1 (better reliability 3.3 and8.3) and later
• Data generated can be permanently stored in user libraries
Slide 42
TDM• Database Manipulation
– Creation of library files
– Loading the various pure component and mixture databases
• Property Regression
– Converting data into correlated forms for databank storage for use insimulation
• VLE Regression
–Available in TDM 3.0; PRO/II 8.1 (better reliability 3.3 and8.3) and later
• Data generated can be permanently stored in user libraries
Additional Background - ToolsKeyword files
• UPSTREAM UNIT
–Matching data for production streams
• OPTIMIZERS and CONTROLLERS and Calculator
–Can be used as mathematical solvers
• Output file
– Interaction Reprint
Slide 43
Keyword files
• UPSTREAM UNIT
–Matching data for production streams
• OPTIMIZERS and CONTROLLERS and Calculator
–Can be used as mathematical solvers
• Output file
– Interaction Reprint
Kij Input reprint, Important!
Slide 44
Default after 8.3!
Review and Summarize
I strongly recommend these books!
© 2013 Invensys. All Rights Reserved. The names, logos, and taglines identifying the products and services of Invensys are proprietary marks of Invensys or its subsidiaries.All third party trademarks and service marks are the proprietary marks of their respective owners.
Slide 46
Related Documents
