3P04 Tutorial 7 SingleLoop 2008

7/30/2019 3P04 Tutorial 7 SingleLoop 2008

1/19

Chemical Engineering 3P04

Process Control

Tutorial # 7

Learning goals

1. Experience with a single-loop controller

2. Answering some questions from PC-Education


2/19

An Introductory Experience with Feedback Control

We use many feedback controllers for a typical process

Each controller has one measurement and onemanipulated variable

Feed

Methane

Ethane (LK)

Propane

ButanePentane

Vapor

product

Liquid

product

Process

fluid

Steam

FC-1

F2 F3

T1 T2

T3

T5

TC-6 PC-1

LC-1

AC-1

L. Key

PAH

LAL

LAH


3/19

control display

and computer

A

v1

v2

4-20 mA

4-20 mA3-15 psi

T

Heating medium

Product

composition


Lets look at one example feedback loop


4/19

A

v1

v23-15 psi

T

Heating medium

Product

composition


Our first task is to learn about the dynamics. Wecan apply the process reaction curve experiment.


5/19

0 10 20 30 40 50 60 70 80 90 100-0.5

0

0.5

1

1.5

2

DYNAMIC SIMULATION, Sloop plots deviation variables

Time

ControlledVariable

0 10 20 30 40 50 60 70 80 90 1000

0.5

1

1.5

Time

ManipulatedVariab

le

%open

Molefraction

minute



6/19


Observations from the experiment.

An increase in valve opening increases the

concentration. The process gain is positive. Time to steady-state is about 25 minutes.

Dead time is about 6 minutes and time constantis about 6 minutes.

The process is stable and overdamped.

The measurement has noise


7/19

A

v1

v23-15 psi

T

Heating medium

Product

composition


The computer has a defined algorithm using the feedback

measurement to determine the next value of themanipulated variable

Calculations are done every 1/3 second

The engineer enters the appropriate parameters


8/19

0 10 20 30 40 50 60 70 80 90 100-0.5

0

0.5

1

1.5

S-LOOP plots deviation variables (IAE = 108.3894)

Time

ControlledVariable

0 10 20 30 40 50 60 70 80 90 100

-1

-0.5

0

0.5

1

Time

ManipulatedVariab

le

%open

M

olefraction

minuteKC = 0(Controller off)


Parameter sets the

aggressiveness

of the controller.


9/19

0 10 20 30 40 50 60 70 80 90 100-0.2

0

0.2

0.4

0.6

0.8

1

1.2 S-LOOP plots deviation variables (IAE = 30.8484)

Time

C

ontrolledVariable

0 10 20 30 40 50 60 70 80 90 100

-1.5

-1

-0.5

0

0.5

Time

ManipulatedVariable

%open

M

olefraction

minuteKC = 0.3

Parameter sets the

aggressiveness

of the controller.



10/19

0 10 20 30 40 50 60 70 80 90 100-0.4

-0.2

0

0.2

0.4

0.6

0.8

1


Time

ControlledVariable

0 10 20 30 40 50 60 70 80 90 100

-2

-1.5

-1

-0.5

0

0.5

Time

ManipulatedVariab

le

%open

M

olefraction

minuteKC = 0.8

Parameter sets the

aggressiveness

of the controller.



11/19

0 10 20 30 40 50 60 70 80 90 100-1

-0.5

0

0.5

1


Time

ControlledVariable

0 10 20 30 40 50 60 70 80 90 100

-2.5

-2

-1.5

-1

-0.5

0

0.5

Time

ManipulatedVariable

%open

M

olefraction

minuteKC = 1.2

Parameter sets the

aggressiveness

of the controller.



12/19

0 10 20 30 40 50 60 70 80 90 100-2

-1

0

1

2

3


Time

ControlledVariable

0 10 20 30 40 50 60 70 80 90 100-8

-6

-4

-2

0

2

4

Time

ManipulatedVariab

le

%open

M

olefraction

minuteKC = 1.8

Parameter sets the

aggressiveness

of the controller.



13/19


Observations from the samples of feedbackperformance.

Feedback that is too weak takes a long time tocompensate a disturbance

Feedback that is too strong gives poorbehavior and can result in instability

Feedback can result in underdamped behavior

for a process that is overdamped.

The process dynamics have the dominant effecton control performance.


14/19


Observations from the samples of feedbackperformance.

Even the best behavior is not perfect. Theprocess must be upset before feedback takesaction.

The valve must be moved to a different steady-

state to compensate for a disturbance. We movethe disturbance from the important controlledvariable to the less important manipulatedvariable.


15/19

******************************************************************

* S_LOOP: SINGLE LOOP CONTROL SYSTEM ANALYSIS ** MODIFY FEEDBACK PROCESS PARAMETERS, CV(s)/MV(s) = Gp(s) ** ** Kp (taulead(s)+1) exp (- theta(s)) ** Gp(s) = ------------------------------------------------- ** (tau1(s)+1) (tau2(s)+1) (tau3^2(s)+2(tau3)xi(s)+1) ** ** ** Measurement noise can be added to the dynamic simulation *

* (not the frequency response calculations) by setting kn ** greater than 0. ** ** noise standard deviation ~= 1.4 * kn *******************************************************************SELECT THE APPROPRIATE MENU ITEMMODIFY...

PRESENT VALUES1) Process Gain, Kp 1.0000

2) Dead time, theta 5.003) Time Constant, taulead (numerator) 0.004) Time Constant, tau1 (lag) 2.005) Time Constant, tau2 5.006) Time Constant, tau3 0.007) Damping Coef., xi 1.00

8) measurement noise magnitude, kn 0.059) Return to main menuEnter the desired selection:

S_LOOP parameters for this example


16/19


17/19

************************************************************** S_LOOP: SINGLE LOOP CONTROL SYSTEM ANALYSIS ** MODIFY PID CONTROLLER TUNING CONSTANTS ** /t ** MV(t) = Kc ( E(t) + (1/TI)| E(t) dt - Td (dCV(t)/dt) )** 0/ ** ** MV(s) = Kc ( E(s) + E(s)/(TI s) - (Td s) CV(s) ) *

* ** ** Hints: 1) Set Kc = 0 to turn controller off ** 2) Set TI = 0 to turn integral mode off ** 3) Units ** Kc has units of 1/Kp ** TI and Td have units of theta and tau **************************************************************SELECT THE APPROPRIATE MENU ITEM

MODIFY...PRESENT VALUES

1) Controller Gain, Kc 1.802) Integral Time, TI 7.003) Derivative Time, Td 0.004) Return to main menu

Enter the desired selection:



18/19

*************************************************************

* S_LOOP: SINGLE LOOP CONTROL SYSTEM DYNAMIC SIMULATION ** ** Hints: 1) All inputs are steps at 5% of total time ** 2) If Entry 5 is non-zero, PID controller ** is turned off ** 3) Select Time step (delta t) to approximate ** a) for the smallest time constant, ** (delta t) / tau = 0.05 ** b) dead time/ (delta t) = integer *

* 4) Usually, (total time) / (time step) < 1000 ** 5) Digital controller exec time integer of ** simulation delta time **************************************************************SELECT THE APPROPRIATE MENU ITEM

PRESENT VALUES1) Total simulation time 99.502) Time step for simulation 0.200

3) Set point change 0.004) Disturbance change 1.305) Process reaction curve MV input 0.00

6) Select continuous/digital controller, currently continuous(Controller executed every simulation time step)

7) Execute dynamic simulation8) Return to main menu

Enter the desired selection:



19/19

Reinforcing PID Controller Concepts

Lets Perform the following exercises from PC-Education

Interactive Learning Modules

8.58.9 and 8.13

Tutorial

8.5

(Problems with solutionsit doesnt get any better than this!)

3P04 Tutorial 7 SingleLoop 2008

Documents