Lab 4 - Time Domain Analysis
Lab 4 - Time Domain Analysis
EE312 Control Systems Engineering
Lab 4: Time‐domain Analysis of Second‐order System
Objective:
The purpose of this lab is to study the time‐domain performance of a second‐order feedback control system (Googol
Technology GSMT 2000 series DC Servo Trainer). Specifically, the students will be able to:
Build a second‐order system model in Simulink.
Analyse the time‐domain performance of a second‐order system model using Matlab/Simulink.
Analyse the time‐domain performance of a real (physical) second‐order system via data collection using
Simulink.
Compare the time‐domain performance of the model and real (physical) system.
Duration
1 x 3 hour lab session
Equipment
1. GSMT2014 DC servo system control platform;
2. PC and MATLAB platform;
Introduction & Theory
This experiment is adapted from the laboratory manual for DC Servo System Control Platform GSMT Series.
Specifically, it is derived from Chapter VII – Time‐domain Analysis of Second‐order System.
The GSTM2014 experiment platform is the DC servo motor control system of a double motor based on the high‐
performance motion controller GT400 and the intelligent servo drive. The high‐performance motion controller
GT400 in the GSTM2014 enables real‐time control experiments under MATLAB/Simulink, and the covering topics
such as system modeling and stability analysis and time‐domain analysis of a second‐order system.
The object, the DC servo motor with revolving speed as the input, then the controller of
K ( 0 . 052 s 1)
G c ( s ) , and K, the open‐loop gain, are adopted to construct the unit negative feedback
s
closed‐loop system.
The mathematical model for the object is:
nc (s) 1 1
G 0 ( s )
nu (s) 0 . 052 s 1 0 . 12 s 1
The transfer function for open loop is:
K ( 0 . 052 s 1) 1 1 K
G ( s ) G c ( s ) G 0 ( s )
s 0 . 052 s 1 0 . 12 s 1 s ( 0 . 12 s 1)
Fig. 1 C
Classic structuural diagram of second‐order system
Assumee that the typ
pical second‐o order system has its structtures listed ass above.
T0 1
Where,, when ; T1 01 . 12 ; ; K2 1
the transferr function of tthe open loop is:
K1 K K
G (s)
2
T 0 s ( T1 s 1 ) s ( 0 . 12 s 1 )
。
K 1 K2
open‐loop gaain is K
Where, the o K
1
T0
The cllosed‐loop traansfer functio
on is:
2
K
( s ) n
2
T1 s s K s 2 n s n
2 2
Where,
K 1 1
n (1)
T1 2 K 1 T1
(1) Wheen 0 1 , where dam
mping condittion occurs, tthe step response of the ssecond‐orderr system is a
kin
nd of dying oscillation as sshown in Curvve 1 of Fig. 2..
e
n T
C ( t ) 1 sin( d t ) (t 0 )
1
2
(2)
1
2
1
tg
Where d n 1
2
The Equ
uation (2) maay be used to calculate thee derivation o
of time. Set th
he value as 0,, then:
t
p
n 1
2
d
(3)
C ( t p ) C ( t )
p
C (t )
The overshoot , and
1
2
e
p
(4)
The 2% settling time is t s , and approximates to 4 times of system's time constant 1 ( n ) under 2%
permitted error range, i.e.:
4
ts
n
(5)
(2) When 1 , the critical damp occurs, then the step response of the system is the exponential
curve as Curve 2 in Fig. 2
nt
C (t ) 1 e (1 n t ) (t 0 )
Set the output as 0.98, then the t s may be obtained.
When 1 , i.e. over damping occurs, the step response of the system is the exponential curve as Curve 3
in Fig. 2
s1 t s 2t
n e e
C (t ) 1
t 0
1 s1
2
2 s2
Where,
s 1
2
1 n s2 2
1 n
When is far larger than 1, the s1 influence may be ignored, then
2
1 n t
C (t ) 1 e
t 0
Then the settling time ts is approximate to:
4
t s (6)
2
1 n
C(t)
1
0.02
3
2
t
tp ts
Fig. 2 Transient Response of Second‐order System Step Input
Procedure
The procedure consists of two parts. In Part 1 you will build a model of the second order system in Simulink and analyse its
performance. For Part 2 you will connect the GSMT 2014 model to the PC and collect step response data for analysis using
Simulink.
1. Build second‐order system simulation program in Simulink as shown below, then change the value K 1
into 8.3333, 4.1679, 2.0883 and 1.2327 according to the mathematical model, and calculate overshoot
2) Click on on the upper left of window to create a “Model” window;
3) Open “Simulink\ Continuous” window in the window of “Simulink Library Browse”, shown as follows:
4) Drag “Transfer Fcn” module into the “untitled” window created just now;
5) Double click on the “Transfer Fcn” module, open the following window and the parameter setting is
shown in the following figure;
6) Right‐click on “Transfer Fcn” module and set “Background Color” as Cyan.
7) Copy another “Transfer Fcn” module, double‐click to set parameters shown as the following figure and
set its background color as Green.
8) Drag a “Scope” from “Simulink\Sinks” into “untitled” window;
9) Open ”Scope”, click on on upper left corner, tick “Save data to workspace” and customize the
“Variable name” while “Format” is selected as Array.
10) Drag a “Sum” from “Simulink\ Commonly Used Blocks” into “untitled” window;
11) Double click on the “Sum” module, open the following window and the feedback setting is shown in
the following figure;
12) Drag a “Step” from “Simulink\ Sources” into “untitled” window and set Final Value as 2000;
13) Connect the five modules according to the following figure and save it as
“TwoOrderSystem_Simulation” with the default format of mdl.
8.3333 1
s 0.12s+1
Step T ransfer Fcn1 T ransfer Fcn Scope
14) Click on Button “ ”, and double‐click on Scope module to get system simulation curve. Calculate
tp
overshoot , peak time
t
and settling time s , and then fill them into the experiment results
sheet.
15) Double‐click on “Transfer Fcn1” and change the numerator to 4.1679.
16) Click on Button “ ”, and double‐click on Scope module to get system simulation curve. Calculate
tp
overshoot , peak time
t
and settling time s , and then fill them into the experiment results
sheet.
17) Double‐click on “Transfer Fcn1” and change the numerator to 2.0883.
18) Click on Button “ ”, and double‐click on Scope module to get system simulation curve. Calculate
tp
overshoot , peak time
t
and settling time s , and then fill them into the experiment results
sheet.
19) Double‐click on “Transfer Fcn1” and change the numerator to 1.2327.
20) Click on Button “ ”, and double‐click on Scope module to get system simulation curve. Calculate
tp
overshoot , peak time
t
and settling time s , and then fill them into the experiment results
sheet.
2. Real control
1) Turn on the power button on the electrical control box of DC servo system control platform.
2) Open File “TwoOrderSystem_Control.mdl” in MATLAB/Current Folder, and the real control page shown
as follows will pop up.
Ini ti al i ze
GT 400-SV
Ini ti al i ze
1/ s
8.3333 RpmIn Motor RpmOut
Control
Step Scope
Gai n
Real Control
3) Double‐click to open Real Control module, aGnadi n fill the τ value from Experiment VI into Gain 2,
which is τ=0.052 .
Scope1
Scope2
0.052 du/dt
Gain2 Derivative1
pul s/0.2ms
1 1 1/3000*40 Vel
SetAccVel
V Add Saturation
Gain3
Gain1 Acc
SetAcc
100
Constant
4) Select “Simulation/Configuration Parameters” and the following window will pop up. Click on “Solver” in
attribute‐tree on the left and set “Type” as Fixed‐step.; set size as 0.01; and set “Solver as “ode1
(Euler)”.
5) Click on “ ” compiling program, and there will be prompting message in Command Window
MATLAB after finishing compiling program (There is no need to conduct this step after compiling
program once if the control page structure is not modified. ):
6) Click on “ ” to connect program, and there will be slight sound from the connecting of relay in
electrical control box that can be heard.
7) Click on “ ” to run the program and the motor starts rotating. After its running of about 10 seconds,
and then click on to stop the program.
8) Double‐click to open oscillograph “Scope”, observe the DC servo motor rotation speed response curve
tp
when adding 2000rpm step signal, measure and record the overshoot , peak time and
ts
settling time and fill them into the results sheet.
9) Open”Gain” module and set the value as 4.1679.
10) Click on “ ” to connect program, and there will be slight sound from the connecting of relay in
electrical control box that can be heard.
11) Click on “ ” to run the program and the motor starts rotating. After its running of about 10 seconds,
and then click on to stop the program.
12) Double‐click to open oscillograph “Scope”, observe the DC servo motor rotation speed response curve
when adding 2000rpm step signal, measure and record the overshoot , peak time t p and
settling time ts and fill them into the results sheet.
13) Open”Gain” module and set the value as 2.0883.
14) Click on “ ” to connect program, and there will be slight sound from the connecting of relay in
electrical control box that can be heard.
15) Click on “ ” to run the program and the motor starts rotating. After its running of about 10 seconds,
and then click on to stop the program.
16) Double‐click to open oscillograph “Scope”, observe the DC servo motor rotation speed response curve
tp
when adding 2000rpm step signal, measure and record the overshoot , peak time and
ts
settling time and fill them into the results sheet.
17) Open”Gain” module and set the value as 1.2327.
18) Click on “ ” to connect program, and there will be slight sound from the connecting of relay in
electrical control box that can be heard.
19) Click on “ ” to run the program and the motor starts rotating. After running for about 10 seconds,
and then click on to stop the program.
20) Double‐click to open oscillograph “Scope”, observe the DC servo motor rotation speed response curve
tp
when adding 2000rpm step signal, measure and record the overshoot , peak time and
ts
settling time and fill them into the results sheet.
Experiment Results (Results sheet to be handed in before leaving lab)
1. Copy your Simulink step response graphs of each of the systems tested in Parts 1 and 2 into a Microsoft Word
document.
2. Calculate the theoretical value of the second‐order system performance index in reference of the
parameters given before experiment, and read the performance indexes presented in the simulation
diagram to fill the result into the table below.
3. Observe the collected data in the real (physical) control system to calculate the performance index, and fill
the result into the table below
3. Comment on any similarities or differences in the simulated and real (physical) control system performance
indices.
C ( tp ) C ( ) (%) tp ( s ) ts ( s )
K n Simulat Measur Simula Measu Simulat Measur Simulat Measu Simulat Measu
ed ed ted red ed ed ed red ed red
value value value value value value value value value value
8.3333 8.3333 0.5
4.1679 5.8934 0.707
2.0883 4.1667 1
1.2327 3.2051 1.3