Aspen Plus & Dynamic Workshop (Step by Step)

Aspen Plus & Aspen Dynamic Workshop Driven by Innovation

By: Dinie Muhammad

Presentation Outline • Part 1: Introduction to Aspen Plus • Introduction to AspenONE • Introduction to Flowsheet simulation

• What is Aspen Plus? • What Aspen Plus can do? • Aspen Plus extension- Aspen Dynamic • Steady state and Dynamic model dilemma • How Aspen can help me with my research?

• Part 2: Before starting with Aspen Plus • Process “know how” • Process Analysis

• Property Method

D. Muhammad & AspenTech, 2013

2

Presentation Outline • Part 3: Getting Started with Aspen Plus • Distillation column design • Aspen Analysis 

Binary Analysis  Azeotrope Analysis  Design Specs  Sensitivity Analysis  Optimization • Part 4: From Aspen Plus to Aspen Dynamic • Part 5: Aspen Dynamic with Matlab


3

PART 1: INTRODUCTION TO ASPEN


4

Introduction to AspenONE • Developed by AspenTech Inc. • Integrated simulation software to implement best practices for: Process design and modelling Optimization engineering Production management Supply chain operation Advanced process control


5

General Simulation Problem What is the composition of stream PRODUCT? RECYCLE

REACTOR COOL FEED REAC-OUT

To solve this problem, we need:

COOL-OUT

SEP

PRODUCT

• Material balances • Energy balances D. Muhammad & AspenTech, 2013

6

Flowsheet Simulation What is flowsheet simulation? Use of a computer program to quantitatively model the characteristic equations of a chemical process Uses underlying physical relationships • Mass and energy balance

• Equilibrium relationships • Rate correlations (reaction and mass/heat transfer)

Predicts • Stream flowrate, compositions, and properties • Operating conditions • Equipment sizes


7

Flowsheet simulation


8

Approaches to Flowsheet Simulation Sequential Modular • Each unit operation block is solved in a certain sequence • Aspen Plus is a sequential modular simulation program

Equation Oriented • All equations are solved simultaneously • Aspen Custom Modeler (formerly SPEEDUP) is an equation oriented

simulation program

Combination • Aspen

Dynamics (formerly DynaPLUS) uses the Aspen Plus sequential modular approach to initialize the steady state simulation and the Aspen Custom Modeler (formerly SPEEDUP) equation oriented approach to solve the dynamic simulation


9

Sequential-Modular Approach D. Muhammad & AspenTech, 2013

Equation Oriented Approach 10

Advantage of Simulation Reduces plant design time • Allows designer to quickly test various plant configurations

Helps improve current process • Answers “what if” questions • Determines optimal process conditions within given constraints • Assists in locating the constraining parts of a process

(debottlenecking)


11

Good Flowsheeting Practice • Build large flowsheets a few blocks at a time. This facilitates troubleshooting if errors occur.

• Ensure flowsheet inputs are reasonable. • Check that results are consistent and realistic.


12

What is Aspen Plus? • Steady state computer-aided chemical process simulation tool


13

Aspen Plus Inputs Choose Thermodynamic Models Specify Chemical Components

Process Flowsheet Design


Specify Feed Conditions

Aspen Plus Process Simulation Model Inputs

Specify Operating Conditions

14

What Aspen Plus can do? • Flowsheet (default): process simulation (SA and optimization) • Data Regression: fitting data to existing models in Aspen

• Property Display: show properties of a components in Aspen Plus’s database • Property Analysis: estimating physical and thermodynamic properties • Assay Data Analysis: analyze assay data (petroleum application) • Property Plus: prepare property package for Aspen Custom Modeler


15

Aspen Plus in Process Design & Development


16

Aspen Plus Extension: Aspen Dynamic • Dynamic modeling tool for plant operations and process design • Enables users to study and understand the dynamics of real plant operations • Exported from Aspen Plus steady state model


17

Aspen Dynamic Overview


18

Adding Dynamic Data Data is required to calculate the following: • Vessel geometry (required for vessel volume)

• Vessel initial filling (used for starting liquid holdup) • Process heat-transfer method • Equipment heat transfer options  Equipment heat capacity  Environmental heat transfer


19

Steady state vs. Dynamic dilemma Steady state • All properties are steady (not changing over time). • Can be used to study different steady state conditions for a specific range of properties either at operating conditions or offdesign conditions.


Dynamic • Ability to model the time varying behaviour of a system (changing over time) • Used to analyse the dynamic behaviour (response) of complex systems.

20

Advantages of Steady State Simulation • Immediate answers to system condition variation • Determine results at specific conditions

• Quick what if in design, sensitivity and optimization studies


21

Advantages of Dynamic Simulation • Determine behaviour of plant/system over complete operating range: start up, shut down, accident scenarios, transition between different states and disturbances occurrence (what if –behaviour)

• Can identify in advance if the operating problems occurred • Facilitate the design for control and optimization of process components to ensure optimum system behaviour, even during off design and transient behaviour • Design and commission control systems using simulations and just fine tune during actual installations • Dynamic integrated simulations can help to identify bottlenecks, inefficiencies and safety risks that are not identifiable with steady-state or segregated simulation D. Muhammad & AspenTech, 2013

22

Application for SS and Dynamic Simulation

Mcmillan, G. K. (2006). Modeling and Simulation of Processes. In "Process Control And Optimization" (B. G. Lipták, ed.), Vol. 2. CRC Press, Boca Raton, FL.


23

How Aspen can help me with my research? • Another option for first principle model (FPM) • Simulation and validation of complex chemical process

•Sensitivity analysis and optimization study of process • Study nonlinearity and multiplicity behavior in process • Using Aspen Dynamic & Matlab Simulink for control scheme design


24

PART 2: BEFORE STARTING WITH ASPEN PLUS


25

Process “know how” • Aspen Plus is not a magic box • All the process inputs (e.g. sizing and process condition) must based on facts or heuristic justification • A preliminary study of process design in recommended


26

Process Analysis • Used to generate simple property diagrams to validate physical property models and data • Understand the behavior of the process • Diagram Types: Pure component, e.g. Vapor pressure vs. temperature Binary, e.g. TXY, PXY

Ternary residue maps

• Select Analysis from the Tools menu to start Analysis


27

Aspen Property Method • A collection of thermodynamic models and methods used to calculate physical properties. • Choice of model types depends on degree of non-ideal behavior and operating conditions • Users can modify existing Property Methods or create new ones


28

Case Study - Acetone Recovery • Correct choice of physical property models and accurate physical property parameters are essential for obtaining accurate simulation results.


29

Ideal vs. Non-Ideal Behavior What do we mean by ideal behavior? • Ideal Gas law and Raoult’s law

Which systems behave as ideal? • Non-polar components of similar size and shape

What controls degree of non-ideality? • Molecular interactions

e.g. Polarity, size and shape of the molecules How can we study the degree of non-ideality of a system? • Property plots (e.g. TXY & XY)


30

Comparison of EOS and Activity Models


31

Common Property Methods Equation of State Property Methods • PENG-ROB • RK-SOAVE

Activity Coefficient Property Methods • NRTL • UNIFAC • UNIQUAC • WILSON


32

Choosing a Property Method - Review

References: Aspen Plus User Guide, Chapter 7, Physical Property Methods, gives similar, more detailed guidelines for choosing a property Method.


33

PART 3: GETTING STARTED WITH ASPEN PLUS


34

Aspen User Interface Run ID

Title Bar Menu Bar

Next Button Tool Bar

Select Mode button


Model Library

Model Menu Tabs

Status Area Process Flowsheet Window

35

Case Study Design a distillation process to separate isobutane and propane so that the impurity target in distillate is 2 wt% and in bottom is 1 wt%

Feed: Propane (40%) Isobutane (60%) Flowrate: 100 kg/h Temperature: 322 K (48.85’C) Pressure: ? Feed at Stage 16 Reflux ratio = 2 Number of Stages = 32 (reboiler + sump) Number of Trays = 30 D. Muhammad & AspenTech, 2013

36

Overview of case study C3 0.98 wt% iC4 0.02 wt% C3 0.4 wt% iC4 0.6 wt%

C3 0.01 wt% iC4 0.99 wt% D. Muhammad & AspenTech, 2013

37

How to begin? Develop the distillation column system Specify the C3 and iC4 in component selection Choose a suitable property method

Define feed condition

Specify a reasonable operating condition

Run and check the results D. Muhammad & AspenTech, 2013

38

Columns - Shortcut


39

Columns - Rigorous


40

Develop the distillation column system Valve (pressure changer library)

Pump (pressure changer library) Distillation column – RadFrac (separator library) D. Muhammad & AspenTech, 2013

41

Connect all the blocks

Connect all the red input and output (primary stream) Select material stream to insert stream in the flowsheet D. Muhammad & AspenTech, 2013

42

A complete distillation system P12 FEED

V12

DIST

C1

V1 Rename all the blocks and streams

P11 BOTM

V11

Click the NEXT button and this dialog menu will appeared. Click OK to proceed. D. Muhammad & AspenTech, 2013

43

Fill the specification menu

Select unit measurement D. Muhammad & AspenTech, 2013

Note: You can also use your own set of unit by using Unit-Sets option under the Setup Menu

44

Edit Report Options


Click the NEXT button

45

Specify the component

Click the NEXT button D. Muhammad & AspenTech, 2013

Use the Find button to search the components 46

Select the property method

Select Chao-Seader property method


47

Define the FEED stream


48

How to calculate the pressure in FEED? • Cooling water at condenser is expected to be at 305 K (31.85’C) • Heuristic temperature different for heat transfer in condenser is 20 K

• Therefore, the reflux drum temperature is ~ 325 K • Vapor pressure for C3 at 325 K is ~ 14 atm • Assume the pressure drop in the V1 is 5 atm • So, FEED stream pressure > 19 atm • In this case, FEED pressure is selected at 20 atm


49

Distillation column setup (Configuration)


50

Distillation column setup (Stream)


51

Distillation column setup (Condenser)

Heuristic pressure drop in column = 0.0068 atm



52

Pump 11 and Pump 12 Setup


Use pressure increase 6 atm for all pump 53

V1 Setup

Use outlet pressure option = 14.2 atm

Choose Liquid-Only D. Muhammad & AspenTech, 2013


54

V12 and V13 Setup

Use Pressure drop option = 3 atm

Choose Liquid-Only D. Muhammad & AspenTech, 2013


55

Run the simulation

Click OK to run the simulation


56

The simulation run complete

Result completed normally


57

Status Indicators


58

Check the results (Stream summary>>Streams) Select the wanted streams

The overall result is still not achieve target 

Adjust to STREAMS D. Muhammad & AspenTech, 2013

59

Redesign: RR = 3 • Operating condition for RR is changed from 2 to 3 • Reinitialize the simulation and Run again

Reinitialize button


60

Check the results (Stream summary>>Streams)

Separation target achieved 


61

Analysis Using Aspen Plus • Binary Analysis – This tool will examine and plot the binary interaction between components. • Azeotrope Analysis – To determine whether the mixture is azeotrope mixture or not • Design Spec - This tool will help the user to achieve the production target by varying the specified operating condition. • Sensitivity Tool – This tool will help the user to analysis the effect of specified operating condition over a certain region towards the production target.

• Optimization – This tool will produce the optimized value for the operating condition in order to achieve the desired production target. This tool will automatically change the selected operating value to an optimized value after Run. D. Muhammad & AspenTech, 2013

62

ANALYSIS: BINARY ANALYSIS


63

Find Binary Analysis Menu

Access the Binary Analysis Menu under Tools Menu

Click OK to continue


64

Binary Analysis Menu

Select type of analysis

Select Unit and list/range for Pressure variation

Select basis component

Property Method

Click GO to start analysis


65

Analysis Result Txy Graph

Use Plot Wizard to plot other type of graphs e.g. xy D. Muhammad & AspenTech, 2013

Full results 66

ANALYSIS: AZEOTROPE ANALYSIS


67

Azeotrope Analysis Menu

Select the menu

In this case, consider a feed of water and isopropane mixture to be analyzed. Here, the property method selected is SRK

Mixture Block D. Muhammad & AspenTech, 2013

68

Menu Click the desired component

Select the Pressure basis

Finally, click the Report option to get the analysis


Select Property method and mixture phase

69

Azeotrope Report

Azeotrope exist! D. Muhammad & AspenTech, 2013

70

The xy graph

azeotrope point

xy graph of water and isopropane mixture (from Binary Analysis) D. Muhammad & AspenTech, 2013

71

ANALYSIS: DESIGN-SPEC


72

Choose the Design-Spec Menu

Design Spec and Vary (below) menu in the explorer

Create new ID D. Muhammad & AspenTech, 2013

73

Design Spec Tab Information • Specification – define the target to be achieve in the simulation e.g. 99% composition in distillate stream • Components – specify the target component • Feed/Product Streams - specify the target component’s stream


74

Specification Tab Select type of target

Specify target value

In this case, a mass purity target of 0.99% is desired



75

Components Tab

Select the target component from available components

Propone is selected as the target component



76

Feed/Product Streams Tab

Specify the target stream from the available streams

Since the C3 product stream is at the top, thus the distillate stream is selected D. Muhammad & AspenTech, 2013


77

Vary Menu: To specify the varying variable for Design-Spec

Vary Menu Create new ID


78

Specification Tab Select the varying variable to be used. Must be a variable from the specified operating conditions

Select a reasonable lower and upper bound



79

Run the simulation

Click OK to start the simulation


80

Check result in Vary Menu Select the Results Tab

The final value of RR to achieve 99%C3 purity is 2.87


81

ANALYSIS: SENSITIVITY STUDY


82

Select: Sensitivity Study

Select the Sensitivity option from Model Analysis Tool


83

Sensitivity Study Tab Information • Define: The user need to define the variable to be used as the production/simulation target. • Vary: Choose the a variable from the specified operating conditions to be varied over selected region. • Tabulated: Choose how the data will be tabulated. Usually, varied operating conditions vs. target value responses


84

Insert new variable

Click New and enter a name for the target variable


85

Select the target variable

In this case, we want to specify the C3 mass concentration in the distillate stream as the Target variable



86

Select the Vary Variable

Use search option to find the RR

In this case, the Reflux ratio (RR) is selected to be the Vary variable. The RR variable can be selected by specify C1 (the column) under Block-Var (Block variables).

Specify range: Lower and Upper boundary. Specify the number of point to be plotted D. Muhammad & AspenTech, 2013

87

Tabulate the variables Click Fill variables button as Aspen will automatically tabulated all the variables.

Click the NEXT button and OK D. Muhammad & AspenTech, 2013

88

Check result

Choose Results. Make sure all the result is completed and converged (blue tick on the explorer)

Results summary for C3 composition by varying RR

Full results is available here under S-1 label


89

How to plot results in Aspen

Select the RR column in results summary

Click Plot from menu bar. Specify as X-axis. Repeat the same procedure for C3 result. Finally, click the Display Plot under the same Plot menu D. Muhammad & AspenTech, 2013

90

The Sensitivity analysis results

The figure show the effects of varying the RR towards C3 composition. Based on the figure, the best RR value to achieve the highest C3 purity would be around RR=4


91

ANALYSIS: OPTIMIZATION


92

Select Optimization Menu

Optimization menu

Click New to create a new ID


93

Define Tab

Click New to define a New optimization value

Enter the target variable name and Click OK


94

Define the Target variable

Specify the Target variable The optimization target variable is C3 mass purity in the distillate stream



95

Objective & Constraints Tab Specify the previously defined variable name in the Define Tab

Select max or min Constraint can also be specified in the Constraint Menu

C3 composition is optimized to find the max purity


96

Vary Tab Specify number of varying variable

Select and specify the varying variable

Specify lower and upper boundary

RR is varied from 0.5 to 5 to find the max mass purity for C3 distillate product



97

Run the simulation

Click OK to start the simulation


98

Check the results: Final C3 composition

Final value shows the max C3 distillate product composition can be achieved within the specified boundary


99

Check the results: New optimized RR value

The optimized RR value in the C1 Results Summary


100

PART 4: FROM ASPEN PLUS TO ASPEN DYNAMIC Steady State to Dynamic Simulation


101

Using the same example: A commonly used heuristic is to set these holdups to allow for 5 min of liquid holdup when the vessel is 50% full, based on the total liquid entering or leaving the vessel (Luyben, 2006) • 100% full = 10 minutes of volume flowrate • From Hydraulic Tab: Reflux drum volume = 0.00800586 m3/min (10min) = 0.0801 m3 Sump volume = 0.00216335 m3/min(10min) = 0.0216 m3

*Please refer to slide18 &19 for explanation on dynamic properties Luyben, W. L. (2006). "Distillation Design and Control using Aspen Simulation," Wiley, New York.


102

From Hydraulic Tab: Stage 1 => Reflux drum volume i.e. sum of Reflux and distillate flowrate


103

From Hydraulic Tab: Stage 32 => sump level i.e. liquid entering reboiler from bottom tray


104

Calculate the vessel geometry

Reflux drum: L = 0.9718m; D = 0.3239 m Sump: L = 0.6279 m; D = 0.2093 m L=length; D=diameter D. Muhammad & AspenTech, 2013

105

Vessel Geometry


106

Entering the dynamic properties

Click this button to enter the dynamic properties



107

Enter the dynamic properties in the column configuration: Reflux drum and Sump Sizing

Enter the calculated Length and Diameter for Reflux Drum and Sump



108

Entering the properties for Hydraulic calculation inside the column

Choose Rigorous Tray Calculation



109

Additional Info • Simple Tray: Using simple tray hydraulics equation relates the liquid flow rate from a tray to the amount of liquid on the tray. Here, the Francis weir equation for a single pass tray is used.

• Rigorous: The pressure drop across the tray is calculated by the same rigorous methods used for the steady-state simulation. The Francis weir equation is used to model the hydraulics based on the number of passes and tray geometry specified in the steady-state simulation.


110

Tray Rating

Since we are using Rigorous Tray Calculation, we need to specify the Tray Rating (so that Aspen Plus can perform the pressure drop calculation along the trays)


111

Specify Tray Rating

Click New and enter any ID number

Select Tray Rating menu under the C1


112

Specify Tray Rating Enter the starting stage = 2 and End stage = 31 (In Aspen Plus; Stage1 = Condenser and Stage 32 = Reboiler) Enter the tray diameter, Tray type, Tray spacing and weir heights

Note: Default value for Tray spacing = 0.6069 m weir heights = 0.05 m


113

Pressure Drop profile

In order for the Aspen Plus to calculate and update the Pressure Drop profile inside the column, this box must be tick

Click the NEXT button and RUN the simulation D. Muhammad & AspenTech, 2013

114

Export to Dynamic (Flow Driven)

Click this icon for export our model into dynamic state (flow driven). A menu will pop up to rename and save the model. Just click OK.


115

Additional Note: Aspen provide two type of dynamic simulation i.e. flow driven and pressure driven. The icon for pressure driven simulation is just next to the flow driven in the menu. In the author experience, flow driven simulation is much simpler to develop compared to the pressure driven. Once the simulation is completed with no error, the simulation is ready to be export to the dynamic states in flow driven. However, for pressure driven, all the pressure inside the streams in steady state model must be control by using pump or valve and its pressure must appropriate. There are also problem (depends) with irregular pressure drop inside the column and inconsistence pressure in feed and recycle stream. Use the Pressure checker icon to check the pressure within the SS model. Refer Process Simulation and Control Using Aspen by AK Jana. Pressure Checker D. Muhammad & AspenTech, 2013

116

Find the saved file .dyn file Click the saved file from previous menu. Generally, the file is saved in the same folder as the SS simulation file


117

Entering Aspen Dynamic (or Custom Modeler)

Choose the state of simulation: Dynamic or Steady-state. Run Initialization at before starting dynamic simulation


If all goes right, you should get this figure. Notice that in Aspen Dynamic, the basic controller is already implemented. These control loops are important to operate the column properly. Click this set of icons to run/pause/rewind (or restart) the simulation

118

Additional Info: • For distillation system, there are 3 major control loop that are essential to operate the column:1.

Top / Condenser Pressure control loop –control energy balance

2.

Reflux drum Level control loop –control mass balance (top)

3.

Sump Level control loop –control mass balance (bottom)


119

See simulation result Run the simulation. Right click top product stream. Select Forms and click TPFmPlot

During running the simulation, this panel will show the latest calculation step


120

Results in real time form

This panel display the mass flowrate, pressure and temperature for the top product stream in real time. Use Zoom Full option for clearer plot.

Although the graph is not steady, notice that the difference (in each parameter) is very small. D. Muhammad & AspenTech, 2013

121

Specify custom parameter (e.g. Propane purity in top product stream)

#2 Name form and choose Plot option

#1 Select Tool in the top menu. Click New Form


#3 The plot figure with no Y axis value

122

Specify parameter (e.g. Propane purity in top Specifycustom specific parameter product stream) #4 Right click top stream and choose Results in the Forms option

#5 From Results Table, drag the highlighted row (Propane purity) into the Y axis of the plot. The final figure should be like the one on the left. Run the simulation in dynamic mode.


123

Specify custom parameter (e.g. Propane purity in top product stream) #6 We can now know the Propane composition in Distillate Stream in real time


124

PART 5: ASPEN DYNAMIC WITH MATLAB SIMULINK


125

Getting Started with Aspen-Matlab • Basically, AspenTech had made a collaboration with Mathworks to develop the AMS simulation system to connect Aspen Dynamic with Matlab Simulink

• However, there might be some compatibility issues regarding Aspen and Matlab version. Please refer to Aspen Help. Based on the author experiences:  Aspen V7.2 compatible with Matlab 2009  Aspen V7.3 compatible with Matlab 2010


126

Use Aspen Dynamic Examples • As an example, we are using the Simulink file in the Aspen Dynamic Examples • Find the Aspen Dynamic instillation folder. Inside the folder, find the Examples folder. Inside the example folder, click the Simulink folder;  C:\Program Files\AspenTech\Aspen Plus Dynamics V7.2\Examples  Click the MCH file (Simulink) as shown below:

Note: MCH is a simulation of extractive distillation of methylcyclohexane and toluene using phenol as an entrainer.


127

The MCH simulation in Simulink

AM-Simulation Block Notice that there are 4 control loops that are controlling the MCH column. Now, input s form the Aspen Dynamic (via AMS Block) is supplied to the controller block. Then, the controller action is computed in Simulink and returned back to the Aspen Dynamic for further action. D. Muhammad & AspenTech, 2013

A step input block act as the disturbance

128

Configure AMSimulation Block Click the AMSimulation Block to open this menu

Use Browse to find the .dynf (Aspen Dynamic) file Click Connect to link with Aspen Dynamic

Input & Output represent the variables that being used in the AMS Block. Input refer to the input that is supplied to the Aspen Dynamic model (e.g. MV or DV). Output refer to the process variable (i.e. PV) that is produced from the model. D. Muhammad & AspenTech, 2013

MCH Model in Aspen Dynamic 129

AMSimulation file Follow the link

Before begin the Aspen-Matlab simulation, it is advised that we copy the AMSimulation file (mfile format) into the current working folder (in Matlab) . The file is generally located inside the AMSystem folder in the Aspen installation folder.


130

Running the simulation Scope Aspen Dynamic

Simulink

Click RUN button in the Simulink to run the simulation D. Muhammad & AspenTech, 2013

131

How they work?

AM-Simulation

Compute and provide controller action (decide the MV)

Matlab Simulink D. Muhammad & AspenTech, 2013

Provide simulation data and result (present the PV)

Aspen Dynamic 132

What happen? • Based on the previous figure (after running the simulation), Matlab Simulink had provided the initial Input (SS or initial value) for the Aspen Dynamic Model. Then, the input is processed (or calculated) by Aspen Dynamic to provide the current process variable (PV) values. The process variables is send back to Simulink environment via AMS Output. • Based on the output that we had selected (in the AMS box), the output will provide the latest PV for Simulink Matlab to calculate its next MV. The new MV is then supplied back to the Aspen Dynamic via AMS Input and so on. • One of the ways to set the initial value for the Aspen Dynamic is by using the unit delay box in Matlab Simulink.


133

Simulation Time • In the author opinion, it is important to synchronize the Aspen Dynamic and Matlab Simulink simulation time. • This can be done via RUN (in the menu bar) >> Run Option or select F9.

Adjust the time units to match both simulation time


134

Simulation model vs. predictive model

u(k)

Simulation Model (Aspen Dynamic)

y(k)

u(k)

Predictive Model

y(k+1)


135

Special Thanks • Assoc. Prof Dr. Norashid Bin Aziz (USM) • Assoc. Prof Dr. Zainal Bin Ahmad (USM)

• Imam Mujahidin Iqbal, Msc (USM) • Process Control Research Group (PCRG) USM


E: [email protected]

136

END OF PRESENTATION


137

Aspen Plus & Dynamic Workshop (Step by Step)

Recommend Documents