PRO/II Training

ECUST
PRO/II
Advanced
Training
Copyright 2004 SIMSCI-ESSCOR All Rights Reserved

PRO/II Training

What
is
the
Power
of
Simulation3?
Who we are

trusted results

Provider of software and services to the

Hydrocarbon and Power industries that allow for
plant start-up, operation and
efficient:
training of operators with

plants and processes

steady state & dynamic
simulation
(PRO/II, HEXTRAN,
VISUAL FLOW,INPLANT
PIPEPHASE, DYNSIM,
TACITE,
iFEED Suite/COMOS)

modeling, dynamic
simulation & control
emulation

design

operate
increased
profitability

optimize

(PRO/II, DATACON, OTS,

DYNSIM, FSIM)

advisory and closed loop of

their processes
(ROMeo, ARPM, Connoisseur,
PRO/II, NETOPT)

focused on simulation
2

PRO/II Training

SimSci-Esscors
Vision
Be the leading provider of software and solutions
for

Simulation & Modeling

Performance Monitoring
Optimization
Operator Training
Data Reconciliation
Integrated Process Engineering
Collaborative Engineering
Use the Best-in Class Technology
& Expertise
Control Checkout
Reactors
Thermodynamics
Intuitive Engineering GUI
Separations
Heat Exchange
Applied
Simulation
Every Engineers
Desktop
Deliver
Performance
for our Customers
Improved
Fluid
Flow Asseton
Optimization
3

PRO/II Training

products

sample applications verticals

sim4me
SIM4ME - Delivering
on our Vision
Hydrocarbon
Power
Pulp & Paper

Process
Design

High Fidelity
OTS

MRA &
ROMeo

Flare System
Design

Decision
Support

PowRx

Well Design/
Nodal Analysis

Engineering
Studies

Oil/Gas
Crude
FCCU
Ethylene

Debottlenecking

PRO/II
HEXTRAN

Oil/Gas
Crude
FCCU

PIPEPHASE
NETOPT

DYNSIM
OTS
FSIM
TACITE

design

operate

VISUAL FLOW

Plant

Lifecycle

ATI/Hyprotech
CANNOT do this
easily with their
current
architecture!

Ethylene
Crude
FCCU
Gas Lift
Optimization

ROMeo
ARPM
MRA
Connoisseur

optimize
Management

PRO/II Training

Application During Plant Lifecycle

SIM4ME

Basic Design
Concept

Revamp

tate
S
y
d
Stea

Engin
eering
Dbs
D

On
line
Op
tim
Ad
iz a
van
tio
ced
n
S im
C
on
u la
tro
tio
n&
l
Pla

nn
ing

Detailed Design
yn
am

ic

Plant
Design

Si
m
ul

at
io
n

ol
r
t
n
Co

m
ste
y
S

Controls

Operator Training

Operation
Commissioning

Construction
5

PRO/II Training

Process Engineering Suite

PRO/II

Process Flowsheet Simulator for Design,

Operational Analysis, and Optimization

HEXTRAN

Heat Exchanger Network Simulator for Design,

Operational Analysis, and Optimization

DATACON

Data Reconciliation Program for

Heat/Mass/Composition balance on plant data

INPLANT

Plant Piping and Utility Systems Flow Simulator

VISUAL FLOW

Flare Network and Regulatory Compliance

Simulator
6

PRO/II Training

PES Features

Enhances productivity in the plant life

cycle
Basic Design
Concept

Revamp

PLANT
LIFE
CYCLE

Detailed Design
Plant
Design

Controls
Operation &
Troubleshooting

Commissioning

Construction
7

PRO/II Training

Integration within PES

Hextran
MS Office

Datacon

PRO/II

Inplant

Complete
Could be done
The Plant

Visual Flow
8

PRO/II Training

Introduction

History of PRO/II

First Generation: 1974

Second Generation: 1979

PRO/II with Provision 4.x

PRO/II with Provision 5.x
version 5.61 March of 2002

Sixth Generation: 2003

Version 3.30 - Spring of 1993

Fifth Generation: 1997

PRO/II Simulation Program

Forth Generation: 1995

Process Simulation Program

Third Generation: 1988

SSI/100 Simulation Program

Over 40 new features

Seven Generation: 2004

Just released in August 2004

PRO/II Training

PES Solution Client Benefits

Reduced process engineering time & cost

Reduced plant capital cost
Reduced plant lifecycle costs
Increased plant operating profits

higher product rates

improved product quality
lower operating costs
more feed flexibility

A valuable tool for experienced process manager

and engineers
10

PRO/II Training

Flowsheets Features
PRO/II is much better for larger flowsheets

No over-specify flowsheet
Recycles estimates not required
Recycle block not required
More option to define sequence
Easier diagnosis of problems since each specification in
linked to a particular unit operation and color indicates
status.

11

PRO/II Training

Distillation Features
Multiple column algorithms to model complex columns
IO, Sure, Chemdist, Liquid-Liquid, Electrolytes, Enhanced IO
Multiple methods for generating initial estimated values
Simple, Conventional, Refinery, Chemdist, Electrolytes
Reactive distillation
robust algorithm
derivative data not required
Tray Hydraulic for rating and design
Volve, Sieve, and Cap structured tray
Sulzer structured packing
Norton random packing
12

PRO/II Training

Reaction Option
Enter reactions in the reaction data section
In Reactor units, select which reactions to use
First Create
a Library of
Reaction Data

Then Select
Reactions
for Each Unit

13

PRO/II Training

Reactor Types
General: (no reactor geometry required)
CONVERSION REACTOR (multiple reactions)
EQUILIBRIUM REACTOR (multiple reactions)

Kinetic: (reactor geometry required)

PLUG FLOW REACTOR (PFR)
CONTINUOUS STIRRED TANK REACTOR (CSTR)

GIBBS: (stoichiometry optional)

Free energy minimization
Kinetics not considered

14

PRO/II Training

Optimizer Features
Optimize based on an objective function
Utilizing tag data values
No needs to have a dynamic calculation
Automatic identification of the best design or
operating conditions from a collection of
alternatives
Frees user from evaluating all possible cases

15

PRO/II Training

User-Added Program Features

UAS/PDTS enhancements

additional function calls

additional subroutines
additional simulation database access
full documentation
supported in PROVISION

16

PRO/II Training

Tag Data Features

Directly access plant historical data
Read tag data from a file
Read tag data from server

PI
ODBC
@aGlance/IT
AIM
Write tag data back to a file

17

PRO/II Training

OLE Features
OLE/COM Automation Layer
documented access to simulation database for most
data
two way link
y simulation data out, design/plant data in

any OLE compliant application

y e.g. MS Office can use VB or VBA

used for Zyqad or Icarus interface

examples available at www.simsci.com

18

PRO/II Training

Spreadsheet Tools

19

PRO/II Training

OLE Automation

20

PRO/II Training
11

Operator Interface

TURBOEXPANDER PLANT

C1
100
10

8
E1
2

E2

2
3X
3

4
7
5

6
6
7

Feed
Flowrate
Pressure
Temperature
Composition
N2
C1
C2
C3
IC4
NC4
IC5
NC5
NC6
NC7

100
1016700.0000 FT3/HR
587.0000 PSIG
120.0000 F

Range

Products
Flowrate
Pressure
Temperature
Composition
N2
C1
C2
C3
IC4
NC4
IC5
NC5
NC6
NC7

7.9100
73.0500
7.6800
5.6900
0.9900
2.4400
0.6900
0.8200
0.4200
0.3100

9
483.7454
125.0000
24.0874
0.0000
0.0086
0.3633
0.3141
0.0548
0.1351
0.0382
0.0454
0.0233
0.0172

11
D1
2195.4231
LB-MOL/HR
161.2292 PSIG
5
157.5738 F
0.0965
0.8896
0.0137
0.0002
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000

X1

T1

10
9

V1

Units
X1 adiabatic efficiency
X1 outlet pressure
C1 adiabatic efficiency
C1 work from X1
T1 top pressure
T1 C1:C2 ratio bottom
E1 HICO
Stream 3 temperature

80.0000
125.0000
75.0000
0.9000
125.0000
0.0150
10.0000
-83.9990

%
PSIG
%

0-100
0-100
0-1

PSIG
0-1
F
F

Units
Reboiler duty
E1 duty
E2 duty
X1 actual work
C1 actual work

2.2889
5.2814
4.9133
392.2247
353.0023

MM BTU/HR
MM BTU/HR
MM BTU/HR
HP
HP

Run Simulation

21

Introduction

PRO/II Training

Simulation in Seven Steps

2

Check
Units
of
Measur
3
e

Build
Flowshe
et

Select
Therm
o
5

Define
Compone
nts

Suppl
y
Strea
m
Data

Provide
Process
Condition
s
7
Run &
View
Results

22

PRO/II Training

Defining the
Components

PRO/II Training

Defining the Components

Component Types
Library component
Petroleum component
User-defined component
Solid component
Polymer component
Ionic component

24

PRO/II Training

Defining the Components

Component Selection

25

PRO/II Training

Defining the Components

PRO/II Component Library

Composite of several databanks
Databank search order

PROCESS (default)
SIMSCI (default)
DIPPR or OLI available as an optional PRO/II add-on

Component selection by list or access name

Pure component data

Fixed properties
Temperature-dependent properties
26

PRO/II Training

Defining the Components

Adding Library Components

27

Defining the Components

PRO/II Training

Component Data Printout

Component name

Phase type

Component type

Nine fixed properties

PROJECT TRAINING
PRO/II INPUT
PROBLEM COMPONENTS
COMPONENT DATA
============================================================================
NO.
--1
2
3
4

COMPONENT NAME
-------------N2
C1
C2
C3

COMP. TYPE
PHASE
MOL. WEIGHT
SPGR
----------- ----------- ----------- ---------LIBRARY
VAP/LIQ
28.013
0.80811
LIBRARY
VAP/LIQ
16.043
0.30000
LIBRARY
VAP/LIQ
30.070
0.35640
LIBRARY
VAP/LIQ
44.097
0.50770

NO. COMPONENT NAME

NBP
CRIT. TEMP. CRIT. PRES. CRIT. VOLM.
F
F
PSIG
GAL/LB-MOL
----------- ----------- ----------- -----------320.440
-232.420
477.619
10.7963
-258.682
-116.680
652.499
11.8628
-127.534
90.140
693.648
17.7343
-43.726
206.006
601.652
24.3247

--1
2
3
4

-------------N2
C1
C2
C3

NO. COMPONENT NAME

--1
2
3
4

-------------N2
C1
C2
C3

ACEN. FACT. HEAT FORM.

G FORM.
BTU/LB-MOL BTU/LB-MOL
----------- ----------- ----------0.04500
0.00
0.00
0.01040
-32066.21
-21726.14
0.09860
-36120.21
-13810.45
0.15290
-44650.04
-10139.64

28

PRO/II Training

Defining the Components

Using DATAPREP
Menu-driven DOS interface
Total access to PRO/II component database
Additional information:

Fixed properties
Data source
Data accuracy
Plots and tables

29

Defining the Components

PRO/II Training

Enthalpy Curve for Water from

DATAPREP
(10E+ 2)
320
Ideal Gas
Curve

ENTHALPY BTU/lbmol

240

Sat urat ed
Vapor Curve

160
Heat of Vaporizat ion
at NBP

80
Solid Curve

Heat of Fusion
at NMP

Crit ical Point

-80
-500

Sat urat ed Liquid Curve

-250

250

500

750

Temperat ure F

30

PRO/II Training

Defining the Components

Petroleum Components
Normal Boiling Point
Gravity
Molecular Weight
At least two of three required

31

PRO/II Training

Defining the Components

User-defined Components
Component Name
Component Properties

32

PRO/II Training

Defining the Components

Component Properties
Fixed properties
Temperature-dependent properties
User Defined and Refinery Inspection properties
Solid properties
Polymer properties
Structure data

33

PRO/II Training

Defining the Components

Component Property Window

34

PRO/II Training

Selecting the
Thermodynamics

PRO/II Training

Selecting the Thermodynamics

Example: Propane-Propylene Splitter

Choice of Thermo strongly effects results!

Thermodynamic Condenser Reflux/Feed

System
Duty
Ratio
Peng-Robinson

-59.6

13.1

Grayson-Streed

-37.3

8.2

36

PRO/II Training

Selecting the Thermodynamics

Thermodynamic Data
Required for all flowsheets
Thermodynamic Property Methods
Transport Property Methods

Required for certain units:

y Column

y Dissolver

y Rigorous heat exchanger

y Depressuring unit

y Pipe

y Output tables

37

PRO/II Training

Selecting the Thermodynamics

Thermodynamic Properties
K-Values
(Mass Balances)

Enthalpies
(Heat Balances)

Entropies
Densities

38

PRO/II Training

Selecting the Thermodynamics

K-Value Calculation Methods

Ideal
Equation of State
Liquid Activity
Generalized Correlations
Special Packages
Electrolytes
Polymers

39

PRO/II Training

Selecting the Thermodynamics

Selecting the Thermodynamic Method

40

PRO/II Training

Selecting the Thermodynamics

Enabling VLLE Calculations

Default

41

PRO/II Training

Selecting the Thermodynamics

Modifications
Very important to choose the correct
thermodynamic method
Even more important to insure that binary
interaction parameters are available

42

PRO/II Training

Selecting the Thermodynamics

Modifications (Cont.)
Advanced Equations of State

Model hydrocarbon behavior

Advanced Alpha forms
Advanced mixing rules
Databank of regressed binary interaction
coefficient

43

PRO/II Training

Selecting the Thermodynamics

Modifications (Cont.)
Liquid Activity Coefficient methods

Model non-ideal behavior

Databank of regressed binaries
Databank of azeotropes
Fill options for binaries

44

PRO/II Training

Selecting the Thermodynamics

Modifications (Cont.)
Generalized Correlation

Typically designed for a specific application

Do a good job for heavier hydrocarbons

45

PRO/II Training

Selecting the Thermodynamics

Modifications (Cont.)
Enthalpy, Entropy and Density

Library correlation for enthalpy

No Library correlation for entropy
Library correlation for density
Rackett parameters in Library

46

PRO/II Training

Selecting the Thermodynamics

Transport Properties
Viscosities
Thermal conductivities
Surface tension
Liquid diffusivity
4 methods: Pure, Petroleum, Trapp, User-defined

47

Selecting the Thermodynamics

PRO/II Training

Calculation with Two Liquid Phases

Water decant option

Rigorous VLLE
calculations

L = HC + W

L1 = HC + W

W = pure water

L2 = W + HC
48

Selecting the Thermodynamics

PRO/II Training

Water Decant Option

Vapor
VLE K-values

Liquid

Water Vapor
Pressure

Pure
Water Solubility Water

49

PRO/II Training

Selecting the Thermodynamics

Rigorous VLLE Calculations

Vapor
VLE K-values

Liquid 1

VLE K-values

LLE K-values

Liquid 2

Must enable two-liquid phase calculations.

50

PRO/II Training

Selecting the Thermodynamics

Hydrocarbon Systems
Refining Processes:

Grayson-Streed: Hydrogen rich systems, Crude tower,

Vacuum unit, Coker fractionator, FCC main fractionator

SRK and PR: Light ends columns, Splitters, Gas

recovery plants, Hydrogen rich systems (SRKM)

SOUR, GPSWATER: Sour water systems

SRKK, SRKM, SRKS, IGS: Use if H/C solubility in
liquid water (VLLE) is important.

51

PRO/II Training

Selecting the Thermodynamics

Hydrocarbon Systems
Gas Processing:

SRK and PR: All types of processing plants, cryogenic

systems

SRKM, PRM, and SRKS: Systems with water,

methanol, and other polar components

GLYCOL: Dehydration with TEG. Improved for

aromatic emissions. Based on SRKM.

AMINE: Natural gas sweetening.

SRKK, IGS, SRKM, SRKS: Use if light gas solubility in
water (VLLE) is important.
52

PRO/II Training

Selecting the Thermodynamics

Online Thermodynamic Help

Reference Manual

Detailed technical reference

Application Guidelines

When to use each method

53

PRO/II Training

Selecting the Thermodynamics

Chemical Systems: Activity coefficient

methods
Non-ideal components

Low to medium pressures

Rely on binary interaction parameters (if missing
will be close to Ideal!)
Missing parameters estimated from structures,
azeotrope composition, mutual solubilities etc.
Used with Henrys Law for non-condensibles
VLLE with some methods
54

Selecting the Thermodynamics

PRO/II Training

Chemical Systems: Activity coefficient

methods

NRTL
UNIQUAC
WILSON
UNIFAC

Two
Liquids?

Binary parameters
in databank?

Yes

Yes

Yes

Yes

No

No

Yes

Estimates non-ideality
from structure

Other methods - see Reference Manual

55

PRO/II Training

Selecting the Thermodynamics

Chemical Systems: Equations of State

SRK-SIMSCI, SRKM, and PRM for polar mixtures
SRK-Hexamer for mixtures involving HF
Can model high-pressures
Also rely on binary interaction parameters
Some binary parameters in databanks for above
methods

56

PRO/II Training

Multicomponent
Distillation

57

Multicomponent Distillation

PRO/II Training

Tray Model
_

Vj yj
_

Lj , Vj Liquid, vapor flowrate

_

Lj-1 xj-1

VDj

Fj

Feed flowrate

Qj

Heater/cooler duty

_ _

Fj X F

xj , yj Liquid, vapor mole frac

Tj Pj

XF

Qj

Feed mole fractions

hj , Hj Liquid, vapor enthalpies

LDj
_

Vj+1 yj+1

Lj xj

Subscript denotes
tray number

Tj , Pj Temperature, pressure
LDj

Liquid Draw rate

VDj

Vapor Draw rate

Overbar denotes
_
component
vectors: e.g., x = (x1, x2,
...xNC)

58

PRO/II Training

Multicomponent Distillation

Tray Numbering
Normally use Theoretical Trays (Stages)
Numbered from top down
Condenser is Stage 1

Even for subcooled condenser

Reboiler is last stage

Thermosiphon adds 2 stages

Convert packing to stages

Rule of Thumb: 2 to 3 feet of packing per stage

59

Multicomponent Distillation

PRO/II Training

Tray Efficiency
Murphree Efficiency = 75%
z

yA
100%
efficient:
step to
equilibrium
curve

xA

75% efficient:
step 3/4 to
equilibrium curve

xA
60

PRO/II Training

Multicomponent Distillation

Other Tray Efficiency Models

Vaporization

yi = ciKixi

Equilibrium

Ks adjusted towards 1.0

Vapor leaving stage not at dew point

Can lead to Mixed Phase Condenser product
Better to use Overall Efficiencies

Theoretical / Actual trays to carry out separation

Use different values in different column zones
Tune from experimental data if possible

61

Multicomponent Distillation

PRO/II Training

Overall Efficiencies
Efficiency increases as components decrease
Efficiency increases as reflux increases

Reflux

Results can be very sensitive to number of trays

High reflux: number of
stages strongly
affects results

Low reflux:
number of stages
is less important

Number of Stages
62

Multicomponent Distillation

PRO/II Training

Typical Overall Tray Efficiencies

SERVICE
Simple Absorbers/Strippers
Reboiled Absorbers/Strippers
Deethanizers
Depropanizers
Debutanizers
Deisobutanizers (Refluxed)
Splitters
C2, C2C3, C3C4or C5

PERCENT
20-30
40-50
60-65
65-75
80-90
85-95
85-95
95-100
90-100

Notes:
1) Assume 65-75% for most columns with reboilers and condensers.
2) At low reflux, split insensitive to number of trays in the model.
3) Pumparounds usually modeled as 2 stages.

63

PRO/II Training

Multicomponent Distillation

All Column Algorithms are Iterative

Want to solve f(x) = 0
Generate a sequence of estimates of solution:
x0, x1, x2, ... xN
Equations are satisfied when x stops changing:
| xN - xN-1 | < 0.00001
xN is regarded as the solution

64

Multicomponent Distillation

PRO/II Training

Convergence of Newtons method ...

Good initial guess
leads to solution

f(X)

n +1

f
= x
x
n

n
f
x
(
)

xn

Solution

0
x0 x1

x2

x*

X
65

PRO/II Training

Multicomponent Distillation

Convergence is not guaranteed!

f(x)

0
x*

X
66

Multicomponent Distillation

PRO/II Training

Convergence is not guaranteed!

f(x)

Periodic

0
x*

X
67

PRO/II Training

Multicomponent Distillation

Convergence is not guaranteed!

f(x)

Bad guess
converges...

But better
guess fails!

0
x*

X
68

PRO/II Training

Multicomponent Distillation

Available Distillation Algorithms in

PRO/II
Inside Out (I/O)
Chemdist
Sure
Liquid-liquid
Enhanced I/O

69

PRO/II Training

Multicomponent Distillation

Inside Out (I/O) Algorithm

Relatively ideal thermodynamics including hydrocarbon

with water decant
Incorporates sidestrippers into column -- No recycle!
Thermosiphon reboilers
Flash zone model
Very forgiving of bad initial estimates
Fast!
No VLLE

70

Multicomponent Distillation

PRO/II Training

I/O
Column
Features

1
2
Heater/Cooler

2 phase
condenser +
water decant

Side Streams

Heat Source/Sink

Multiple Feeds

Side Columns

Pumparounds
N-1

Kettle and
Thermosiphon
N Reboilers

71

PRO/II Training

Multicomponent Distillation

I/O Algorithm Uses Nested Loops

Inner Loop

Simple thermo model (Fast)

Approximate Matrix Inversion (Fast)
Converge enthalpy balance and performance specs

Outer Loop

Updates, checks thermo using rigorous model (Slow)

Checks Bubble Point Criteria
If thermo changing or not bubble point, goto Inner Loop

72

Multicomponent Distillation

PRO/II Training

I/O Algorithm

Outer Loop

Prepare approximate thermo

models for K*(K) and H*L(HL),
and H*V(HV).

Approx. Thermo.
Model

Iteratively solve the column

equations using approximate thermo,
K*(T,P) and H*(T,P).

Inner Loop

x, T, L, V, Q
...
1) Calculate rigorous K(x,T,P),
H(x,T,P).
2) If K and H differ significantly
from
previous iterate, repeat from
Done, solution is: x, T, L, V,
beginning.
Q ...

Convergence
Check

73

Multicomponent Distillation

PRO/II Training

Initial Estimate Generator (IEG)

Generates good initial estimates for all column variables
P1 P2
PN LN
Column
Specs

IEG

You supply column

specs and guesses
for a few variables...

x0 y0 T0
P V0 L0
Q0R Q0C

Solver

IEG calculates
initial estimates for
all column variables...

x* y* T*
P V* L*
Q*R Q*C

Solver (I/O, Chemdist)

converges on solution

74

PRO/II Training

Multicomponent Distillation

Four Types of IEG

SIMPLE (default): Simple columns
Only choice for liquid-liquid extraction
CONVENTIONAL: Works well with most columns
Based on shortcut methods
Strongly dependent on your product rate estimates
REFINING: Complex refinery columns (e.g., Crude,
Vacuum, FCC main fractionator, Coker)
CHEMICAL: Nonideal thermodynamics (e.g., azeotropic
and extractive distillation). Can be slow.

75

PRO/II Training

Multicomponent Distillation

Specifications and Variables

Specifications are constraints to be met by the column
Variables are calculated to meet specifications.
Column always balances equations and unknowns
To impose a specification, you must add a variable,
otherwise equations and unknowns dont balance.
Example: Impose two product specifications by
declaring reboiler & condenser duties as variables.

76

Multicomponent Distillation

PRO/II Training

Column Status at Initialization

Fixed quantities remain at their current values unless you
declare them as variables.
If no specs/variables provided, default status used:
QUANTITY
Overhead and Bottoms Rates
Side Draw Rates
Duties
Feed Rates
Tray Temperatures
Tray Pressures
Vapor and Liquid Rates
Product Properties (e.g. Viscosity)
Tray Vapor or Liquid Properties

STATUS
Calculated
Fixed
Fixed
Fixed
Calculated
Fixed
Calculated
Calculated
Calculated

77

PRO/II Training

Multicomponent Distillation

Improper Specifications
0% methane in crude column bottoms

Infinitely many solutions

300 lb-mole/hr propylene in overhead

No solutions if column feed only 250 mol/hr propylene

98% ethanol product

No solutions if Water-Ethanol Azeotrope present

78

Multicomponent Distillation

PRO/II Training

What is alpha ()?

Length of correction:

Xn+1 = Xn + n n

0< || < 1

Decrease if full step increases error

1 = .7
Unknown 1

2 = 1

X1
X2

3
X3
Solution

Reject: full step

increases error

Unknown 2
79

PRO/II Training

Multicomponent Distillation

You Can Help I/O by Using Damping

Damping reduces iteration step and suppresses
oscillation
Conventional columns:

DAMP = 1.0 (default)

Columns with steam:

DAMP = 0.6 - 0.8

Crude, Vacuum, FCC Main Fractionator

Less-ideal:

DAMP = 0.2 - 0.6

Increase number of allowed iterations

If oscillations persist, use Chemdist

80

PRO/II Training

Multicomponent Distillation

Reboiler Models
Most reboilers can be simulated as:

Kettle
Thermosiphon with Baffles
Thermosiphon without Baffles

81

Multicomponent Distillation

PRO/II Training

Kettle Reboilers
LN-1

Vapor in Equilibrium
with Bottoms
VN-1

VN

LN-2

N-1
Bottom Tray
VN
BTMS
BTMS

LN-1

N
Reboiler

82

Multicomponent Distillation

PRO/II Training

Single Pass (Once Through)

Thermosiphon

Equivalent to a Kettle Reboiler because Bottoms

is in Equilibrium with VN
LN-1

LN-1
VN
VN

Bottom
Sump
BTMS

Bottom
Sump Reboiler
Sump
BTMS

Baffle
83

Multicomponent Distillation

PRO/II Training

Circulating Thermosiphon Adds 2

Stages
Simulate as TS without Baffles

N-2
Bottom Tray

LN-2

VN-1

VN-1

RV
BTMS

RL

Combined
Sump
BTMS

N-1
Combined Sump
RL

RF

LN-2
RV

RF

N
Reboiler

84

Multicomponent Distillation

PRO/II Training

Circulating Thermosiphon
Simulate as TS without baffle
N-2
Bottom Tray

LN-2
VN-1

RV

VN-1

RL
BTMS
Bottom
Sump

BTMS

Reboiler
Sump

N-1
Reboiler Sump
RL

RF

LN-2
RV

RF

N
Reboiler

85

Multicomponent Distillation

PRO/II Training

Preferential Thermosiphon
Simulate as TS with baffle
LN-2

N-2
Bottom Tray

RV

VN-1
RL

VN-1
LO

Bottom
Sump Lo

Bottom
Sump Reboiler
Sump
RF

RV
N-1
Reboiler Sump
RF

BTMS R
L
BTMS

LN-2

N
Reboiler
86

PRO/II Training

Multicomponent Distillation

Tips...
Start simple

Converge water decant thermo before trying VLLE

Converge side draws before trying sidestrippers
Test pumparound duties with side coolers
Remove coolers from pumparounds so all cooling is
taken at condenser. Then add duties to pumparound.

Recovery usually safer than composition specs

Spec reflux ratio and product rate
Spec reflux rate and component recovery

87

PRO/II Training

Multicomponent Distillation

Tips...
If Water condenses in column:

Increase temperature estimates to keep water in vapor

Reduce steam flow:
y 0.1 lb/Gallon bottoms in main column
y 0.1--0.2 lb/Gallon sidestripper product

Pumparounds solve best when you:

Fix rate and duty, calculate return temperature

88

PRO/II Training

Multicomponent Distillation

Tips...
Excess cooling cause
drying above PA return

Specify Tray
Liquid rate

Remedy: Specify liquid

flow above return tray and
calculate pumparound duty

Declare Duty
as a Variable

Eliminate loops whenever

possible

Break thermal recycles

with reference stream
Simulate furnace as
column tray heater

FZ
89

PRO/II Training

Multicomponent Distillation

Tips...
Dont believe your answers until you:

Verify thermodynamic method with expert

Rerun with tighter column and loop tolerances
Rerun with more pseudocomponents
Rerun with different assay characterization method
Check sensitivity to estimated parameters, i.e.,
number of trays
y Example: Add a stage to column. If results change
drastically, then model is very sensitive to this parameter.
y Assess if this is physical reality or model defect.

90

Multicomponent Distillation

PRO/II Training

Distillation Algorithm Selection

Inside/Out (I/O)

CHEMDIST

SURE

Unique
Features

Side and main columns

solved simultaneously

Reactive distillation
VLLE on any tray

Total pumparounds
VLWE on any tray
Water draw any tray

Strengths

Very fast
Insensitive to initial
estimates

Highly non-ideal
systems

Generality: complex
column and thermo

Thermo non-ideality
NO VLLE capability
(VLWE at condenser)

No pumparounds
Side columns solved
as recycles

Slow
Sensitive to initial
guesses

Non-ideal systems
Mechanically simple
columns
VLLE within column

Free water or water

draw on trays other
than condenser
Total pumparounds
or vapor bypass

Limitations

Hydrocarbons
Applicability EOS & slightly nonideal LACT thermo
Interlinked columns

91

Multicomponent Distillation

PRO/II Training

Distillation Algorithm Selection

Liquid-Liquid
Unique
Features

Enhanced I/O

LLE on each stage

Total draws and water

decants off trays

Strengths

Perform liquid-liquid
extraction

Converges when zero

flowrates on trays

Limitations

Thermo must be a
liquid activity method

IEG does not work for

all cases

Liquid-liquid
extraction columns

Same as I/O

Applicability

92

PRO/II Training

Flowsheet
Optimization

93

PRO/II Training

Flowsheet Optimization

Optimization allows...
Automatic identification of the best design or
operating conditions from a collection of
alternatives
Frees you from evaluating all possible cases

94

PRO/II Training

Flowsheet Optimization

Setting Up the Optimizer

Objective function

A result calculated in PRO/II (duty, product recovery,...)

Usually evaluated with a calculator unit operation
Minimize or maximize this value (i.e., maximize profit)
Include all relevant costs

Optimization variables

A fixed input parameter with defined MIN, MAX values

95

PRO/II Training

Flowsheet Optimization

Setting Up the Optimizer

Process constraints (inequality)

Limits on flowsheet values which cannot be violated

Physical limitations on equipment
y Constrain compressor operation to prevent surging
y Constrain column tray flows to prevent flooding

Process specifications (equality)

Additional criteria imposed on optimum solution

y Total cooling water flowrate = 100
y Kerosene product rate = 10000

96

Flowsheet Optimization

PRO/II Training

One Variable Optimization

Value of OVHD [$/lb-mole] is proportional to the
square of its mole fraction C1 and C2
What temperature maximizes profit from OVHD?
Objective: maximize

[ OVHD (YC1 + YC2)2 ]

OVHD

H2O; C1-C6
-60F
900 psia

T=?
30 psia

97

Flowsheet Optimization

PRO/II Training

One Variable Optimization

1000

Objective
Function
Flowrate of C1
and C2 times
Purity of C1 and
C2

800
600

Optimal
Temperature

400
200
0
-150

-110

-50

10

70

110

Flash Temperature

Optimization Variable

98

Flowsheet Optimization

PRO/II Training

Multivariable Optimization
What temperature and pressure maximize profit
from OVHD?
Objective: maximize

[ OVHD (YC1 + YC2)2 ]

OVHD

H2O; C1-C6
-60F
900 psia

T=?
P=?

99

Flowsheet Optimization

PRO/II Training

Multivariable Optimization
Maximum
1400
1200
1000
800
600
400

35

200

25

Pressure

90

Tempera
ture

50

10

5
-30

-70

15
-110

-150

Objective
Function

100

Flowsheet Optimization

PRO/II Training

Optimization with Constraints

Vary temperature and pressure to maximize
flowrate of C1 and C2 in OVHD
The OVHD purity must be at least 90%
OVHD
H2O; C1-C6
-60F
900 psia

YC1 + YC2 > 0.9

T=?
P=?

101

Flowsheet Optimization

PRO/II Training

Optimization with Constraints

35

0
360
25

Constraint

650

Flash
Pressure

Optimization
Variable

860

15

1147

Optimum
(1411)
5

-150

-110

-70

-30

10

50

90

Flash Temperature

Optimization Variable
102

PRO/II Training

Flowsheet Optimization

Analyzing your Results: Shadow

Prices
Indicates the potential benefit of relaxing a limit,
specification, or constraint

Positive:

Increasing the value increases the

objective function

Negative:

Increasing the value decreases the

objective function

Zero:

Constraints and/or limits on optimization

variables (MINI, MAXI) are not active

103

PRO/II Training

Flowsheet Optimization

Reading the Optimizer Summary

Best results

Objective Function
Values of Variables

Optimizer history at each cycle

Values for objective function and variables

Derivatives (Objective Function/Variable)
Shadow Prices

Convergence plots in output report

104

Flowsheet Optimization

PRO/II Training

Optimizer Output
** BEST OBJECTIVE FUNCTION = 1.41158E+03 AT CYCLE NUMBER 6
VARY
INDEX
----1
2

--------- VARIABLE ---------INITIAL VALUE

OPTIMUM VALUE
------------------------1.00000E+01
-4.12426E+01
3.00000E+01
5.00000E+00

- SHADOW PRICES ---CYCLE

1
---------- ----------VARY
1
0.0000E+00
VARY
2
0.0000E+00
CNSTR 1
-2.4824E+03

5
----------0.0000E+00
-3.1750E+01
-1.4458E+03

BEST - 6
----------0.0000E+00
-3.1430E+01
-1.4146E+03

7
----------0.0000E+00
-3.1428E+01
-1.4149E+03

8
----------0.0000E+00
-3.1116E+01
-1.4338E+03

---- VALUES ---CYCLE

1
---------- ----------VARY
1
1.0000E+01
VARY
2
3.0000E+01
CNSTR 1
9.2355E-01
REL ERR
0.00E+00
SUM SQ ERR
0.0000E+00
OBJECTIVE
9.5967E+02

5
-----------3.9238E+01
5.0000E+00
8.9303E-01
-7.74E-03
5.9977E-05
1.4210E+03

BEST - 6
-----------4.1243E+01
5.0000E+00
8.9935E-01
0.00E+00
0.0000E+00
1.4116E+03

7
-----------4.1439E+01
5.0000E+00
8.9995E-01
0.00E+00
0.0000E+00
1.4106E+03

8
-----------4.1454E+01
5.0000E+00
9.0000E-01
0.00E+00
0.0000E+00
1.4106E+03

105

Flowsheet Optimization

PRO/II Training

Shadow Price Examples

1) Where should you send any extra steam?
Stm1

Process

Stm2

Profit

Stm3

2) Which heat exchanger should you clean first?

Gas
T=200F

Gas
T=10F

106

Flowsheet Optimization

PRO/II Training

Solution Technique: Successive

Quadratic
Programming (SQP)
Initialization
Second order method
(with derivatives) to
determine search direction

Quadratic
Programming
Sub problem

First order method

(no derivatives) to check
progress towards the solution

Line Search

Convergence

107

PRO/II Training

Flowsheet Optimization

Convergence
Converging loops requires more intervention
Derivative step sizes are very important
Tolerances of units in loops should be lowered

108

Flowsheet Optimization

PRO/II Training

Convergence
Specifications and constraints are satisfied and

-7

Scaled error below tolerance (10 ) or

Variables stop changing (tol=0.1%) or
Objective function stops changing (tol=0.5%)

Objective Function
Warning: Optimum is
T=100, but any guess
between 50 and 150
satisfies objective test

1005
1000
Tn
995

50

T (C)

100

150

109

Flowsheet Optimization

PRO/II Training

Convergence: Relative Tolerances

Example: want L = 0.99 F
Which specification should you use?

Form 1:
Form 2:

F
L

L/F = 0.99
V/F = 0.01

PRO/II converges this to a relative tolerance

Form 1:

Converges when: | (L/F - 0.99) / 0.99 | <

so L = 0.99F 0.99F

110

Flowsheet Optimization

PRO/II Training

Convergence: Relative Tolerances

Form 2:

Converges when: | (V/F - 0.01) / 0.01 | <

But, V=F-L :
| (1-L/F - 0.01) / 0.01 | <
so L = 0.99F 0.01F

If =0.01 (the default) and F=1000

Form 1:
Form 2:

L=990 9.9
L=990 0.1

Form 2 is much more accurate!

To use Form 2, tighten relative tolerance
111

Flowsheet Optimization

PRO/II Training

Convergence: Compounding of Errors

Example: specify each flash as Ln = 1/2 Ln-1

Exact solution: LN = (1/2)N L0
If each flash specification relative tolerance =
Then worst case LN relative error ~ N
Example: N=5, =1%, then L5 = L0/32 5%

L0

L1

L2

L3

LN-1

LN

112

PRO/II Training

Flowsheet Optimization

Flowsheet Tolerances
Optimization requires flowsheets to be solved
more accurately than for simulation

Tighten tolerances (columns, recycle loops, controllers)

Choose appropriate finite difference steps
Tighter tolerances allow smaller finite difference
steps to be used which is more efficient

Inaccurate flowsheet information may cause

optimizer to fail or converge prematurely

113

Flowsheet Optimization

PRO/II Training

Finite Difference Derivative

dF(xn)/dx = [F(xn +x) - F(xn)] / x
Largest slope

F(x)

Smallest slope
Error Bar
xn

xn+x
114

Flowsheet Optimization

PRO/II Training

Finite Difference Derivative

Smaller step size can worsen derivatives
F(x)

Calculated slope
can be negative!

xn

xn+x
115

Flowsheet Optimization

PRO/II Training

Finite Difference Derivative

Smaller error bars improve derivative calculations
F(x)

xn

xn+x
116

PRO/II Training

Flowsheet Optimization

Recommendations
Solve base case separately - Check results
Tighten flowsheet tolerances for improved
accuracy
Carefully select bounds and constraints to ensure
physically well-defined flowsheet
Select appropriate convergence criteria

117

PRO/II Training

Questions

Getting Started