DEPARTMENT OF MECHANICAL AND AEROSPACE

ENGINEERING

INTERNATIONAL ISLAMIC UNIVERSITY MALAYSIA

MECH 3118 AUTOMOTIVE LAB 3

CONTROL SYSTEM

EXPERIMENT 2:

ANALYSIS OF AN CLOSED-LOOP SYSTEM

NO. NAME MATRIC NO.

MUHAMMAD ARIEF BIN MOHAMMAD
1 1812383
YAZID
1818171
2 RAHMAT APRILIAN BIN MOHD ISMAIL

1
 TABLE OF CONTENT

INTRODUCTION 3
METHODOLOGY 4
RESULT AND DISCUSSION 7
CONCLUSION 11
REFERENCES 12
APPENDIXES 12

2
INTRODUCTION

A closed loop control system is similar to an open loop system. However, it

consists of at least an additional feedback loop of output transducer or also known as
a sensor, which measures the output response and feeds that measurement back
through a feedback path. This feedback mechanism enables the system to self regulate
and maintain a desired set point without human interaction. It can correct the
incoming disturbances by comparing the output response to the input at the summing
junction. The differences are translated into an actuating signal. When there is a
difference between the two responses, the sensor will send an actuating signal to drive
the plant but when there is none, it will do nothing since the desired output has been
achieved already. If the sensor amplifies the input by a unity gain, the actuating signal
is called the error since its value is equal to the actual difference between the input
and the output.

In order to measure the performance of a closed loop system, there are three
typical criteria that can be used. Transient response can be evaluated by measuring the
rise time and the overshoot percentage. Steady state response on the other hand can be
evaluated by looking at the steady state error generated.

Analysis of a closed loop system is essential because since it is self regulating,

we need to ensure it is properly tuned in order to obtain the best performance possible
within range and minimize the error. This can be done by adjusting the gain in the
loop or in the plant (Proportional controller) or designing a controller to compensate
the system. However, these will affect the system’s gain and stability. By changing the
closed loop gain, for example, the system response may change its behavior to
become slower or faster.

3
Objectives :

1. To evaluate the closed loop system performance by measuring the rise time,
the overshoot percentage and the steady state error.

2. To understand and investigate the effect of changing the closed loop gain to
the system’s stability and transient response.

3. To determine a range of closed loop gain that gives out a stable system by
using Routh-Hurwitz criterion.

METHODOLOGY

1. An elevator system with a transfer function of :

1
𝐺(𝑠) = 2
𝑠 +3𝑠+2
was analyzed.

2. The closed loop step response for a gain, K of 1 was obtained by using
Simulink. The rise time, the overshoot percentage and the steady state error
were obtained by referring to the graph in Simulink. The result was recorded
in Table 1.

4
 Figure 1 Systems’ setup in Simulink for Step 3.

Figure 2 Example of finding rise time in Step 2 for K=1.

3. The same criteria were obtained by using proper commands in MATLAB. The
result was recorded in Table 2.

4. Step 2 and 3 were repeated by using different gains, K of 0.1, 10, and 100,
respectively in order to see the effects of varying gains to the transient and
steady state responses.

5
5. For each of the gains above, the closed loop transfer function and poles were
obtained by using proper commands in MATLAB. The poles maps on the
s-plane were obtained as well and the stability of each case was determined.

6. The closed loop transfer function for each case was obtained analytically. The
range of gain, K that makes the system stable was determined by using
Routh-Hurwitz criterion.

7. One value of gain, K in the stable region and one value of gain, K in the
unstable region were chosen and the responses were simulated separately.

Figure 3 Systems’ setup in Simulink for Step 7.

6
RESULT AND DISCUSSION

Figure 4 Step response for different values of gains.

Figure 4 is a graphical representation of figure 1 using the MATLAB Simulink. In the

figure 4, there are 4 lines of different colors that react differently when a set of
closed-loop transfer function were introduced to four different values of gain, K with
one-line colored yellow served as a set point. Depending on the value of gain, K, the
step response produced can be divided into four categories. The first one is
overdamped, represented by the blue line that only rose slightly before reaching its
steady state. The second state is the critically damped represented by the
orange-colored lines. The third one is the underdamped represented by the green lines
and finally the purple line that can be regarded as an undamped system, but it
eventually settled down to its steady state after a certain period of time.

7
 Table 1 Closed loop system’s criteria obtained by referring to the graph in Simulink.

Gain, K 0.1 1 10 100

Steady State Error, ϵ 0.952 0.666 0.166 0.017

Percentage of
0% 0% 21% 60%
Overshoot
Rise Time (s) 2.196 1.57 0.432 0.113

Table 2 Closed loop system’s criteria obtained by using MATLAB commands.

Gain, K 0.1 1 10 100

Steady State Error, ϵ 0.952 0.667 0.167 0.020

Percentage of
0% 0.43% 22.10% 62.20%
Overshoot
Rise Time (s) 2.438 1.579 0.439 0.116

The performance of each of the system with different value of gain, K can be seen
through Table 1 and 2. These two tables actually represented the same set of data. The
thing that differentiates them is the way the data from each table were collected. The
first table’s data were collected directly from the graph produced from the MATLAB
Simulink. While the second table’s data were recorded by using the MATLAB
commands. From the data, it can be seen that the performance of the system varies
depending on the value of K. The higher the value of K, the better the performance as
the rise time and the steady state error become smaller. But one thing to the note is
that the higher the value of K, the percentage of overshoots would also increase.

8
 Table 3 Poles and Closed loop transfer function of different gains.

Gain, K 0.1 1 10 100

-1.887 -1.50 + 0.866i -1.50 + 3.123i -1.50 + 9.989i
Poles
-1.113 -1.50 - 0.866i -1.50 - 3.123i -1.50 - 9.989i
Closed Loop 0.1 1 10 100
Transfer 2
𝑠 +3𝑠+2.1
2
𝑠 +3𝑠+3
2
𝑠 +3𝑠+12
2
𝑠 +3𝑠+102
Function

Figure 5 Poles map of closed loop transfer function of different gains.

With regard to Table 3, it was observed that the poles of closed-loop transfer function
with stated value of gain, K, are on the left side of the imaginary axis, thus prove that
the system is stable. This data can also be seen through Figure 5, where all the plotted
coordinates of the poles are on the negative side of the real axis. But, this still does
not show what value of gain, K can the system take before it become unstable. To find
this, a method called the Routh-Hurwitz criterion was used. From the calculation, it
was discovered that the system would remain stable (poles on the left side) as long as
the value of K is bigger than -2. This statement was backed by Figure 6, which shows
the step response when K= 0 and 50 are stable but when K is stated to -5 or -2, the
step response would shows that the system are not stable.

9
 Table 4 Routh-Hurwitz criterion

s2 1 2+K 0

s1 3 0 0

s0 0 0

For 2+K>0, K>-2.

Figure 6 Step response for different values of stable and unstable gains.

10
CONCLUSION

Throughout this experiment, the measurement of the rise time, percentage overshoot
and the steady state error are successfully obtained and from them, it was discovered
that by changing the value of gain, K, the stability of the system would be affected.
Other than that, the range of the value of K, in which the system would remain stable
was also obtainable by using the Routh-Hurwitz criterion method.

From the recorded data, the given value of kk which are 0.1, 1, 10 and 100 are not
suitable for the elevator system due to its varying reasons. One of them being that the
value of K is either too big that it cause bumping on the elevator trip up and down or
the value being small but the elevator

11
REFERENCES

[1] Nise, N., 2017. Control systems engineering. 6th ed. Pomona: John Wiley &
Sons.
[2] Routh-Hurwitz Criterion Online Calculator https://www.muchen.ca/RHCalc/

APPENDIXES

MATLAB commands

Step 3:

G = tf([1],[1 3 2]);
T1= feedback(0.1*G,1);
T2= feedback(1*G,1);
T3= feedback(10*G,1);
T4= feedback(100*G,1);
stepinfo(T1)
stepinfo(T2)
stepinfo(T3)
stepinfo(T4)
step(T1,T2,T3,T4)

Step 5:

pole(T1)
pole(T2)
pole(T3)
pole(T4)
pzmap(T1,T2,T3,T4)

12