BKF3791

PROCESS CONTROL & INSTRUMENTATION LAB

SESSION II 2021/2022

SECTION 01

MODULE 4

STUDY ON DYNAMICS OF FIRST ORDER AND SECOND ORDER SYSTEM

LECTURER

DR. LIANG YONG YEOW

GROUP 3 MEMBERS:

NAME STUDENT ID
HARIDHARANI A/ P PERUMAL KA19151
SHARLENE A/P ESOOTHAS KA19154
MERVIN RAJKUMAR A/L NAGARAJAN KA19155
SHUBAHSHINI A/P MAYELVAGANAM KA19162

SUBMISSION DATE

29th APRIL 2022

1.0 ABSTRACT

The main goal of this experiment is to show the properties of first and second order systems
for various input values. Another goal is to demonstrate the dynamic reaction of a first and
second order system to various input signals. A first order system has an output y(t) that is
modelled by a first order differential equation, while a second order system has an output y(t)
that is modelled by a second order differential equation. This experiment is broken into three
parts: A, B, and C. Part A discusses the first order system, whereas Part B describes the second
order system, and Part C describes the system identification problems 1 and 2. In the first order
system block, the system gain Kp and time constant τp at 10.0 for part A are set, and the step
time are set at 10.0 and the step function value are set to 0.0 to 1.0. Then, by clicking once on
the first order system block, both the system gain Kp (numerator coefficient) and the system
time constant τp (denominator coefficient) are to be set to 10.0. The simulation is run, and the
new steady-state value and the time it takes for the output to reach the new steady-state value
are recorded. The technique for the second order system is similar, except that we must enter
the values of A and B before simulation. Second order results are severely damped,
overdamped, and underdamped. If the system is underdamped, the overshoot, decay ratio,
rising time, and settling time are calculated. The dynamic responses of the first and second
order systems to various input signals have been depicted. For first-order findings, the
simulation, system gain, Kp, and system time constant, τp changes. The simulation results in a
critically-damped graph for all first order. The critically-damped graph for second order results
does not include overshoot, decay ratio, rising time, and period because it only has one peak
and no oscillation. Following that, the simulation moves on to System Identification Problem
1 and System Identification Problem 2. The slope of the initial response, the ∆ output value,
the Kp value, and the τp value may all be calculated and derived from the simulation graph.

2
2.0 METHODOLOGY

TASK A: FIRST ORDER SYSTEM

The First and Second Order Systems button are clicked from the Main Menu. then select
the First Order System button.

The First Order System button were selected

The system gain Kp and time constant τP were set based the table 1 according to the first
order system block. Then, the step time and value of the step function were chosen.

The simulation were started and the record of new steady state value and the length of time
it takes for the output to reach the new steady state was taken. Several points along the
response curve were also taken.

The step 3&4 were repeated using a different set of Kp and τP as shown in Table 1.

Figure 2.1: Procedures of first order system

3
TASK B: SECOND ORDER SYSTEM

The Second Order Systems button are clicked.

The system gain Kp, value of A and B were set based the table 2. Then, the step time and
value of the step function were chosen.

Then, the initial and final value of the step function were chosen.

The simulation were started and observation were done on identifying whether the system
is overdamped, underdamped or critically damped. When the system is underdamped the
overshoot, decay ratio, rise time, settling time and the period of oscillation were analysed

The step 2&3 were repeated using a different set of Kp, value of A and B as shown in Table
2. An analysis were made according to step 4.

Figure 2.2: Procedures of second order system

4
TASK C: DETERMINE PARAMETERS OF FIRST AND SECOND ORDERS SYSTEM

The System Identification Problem 1 button were selected from the Main Menu.

The simulation were run to generate output data that can be used to determine the system
gain (Kp) and the system time constant (τP) by using a step input.

The System Identification Problem 2 button were selected.

The simulation were run to generate output data that can be used to determine the
parameters of second order system by using a step input.

Figure 2.3: Procedures of system identification problem 1&2

5
3.0 RESULT

TASK A: FIRST ORDER SYSTEM

No. Kp τp Time Output Comments

1 10 10 66 9.963 According to Figure xx, when

the Kp and τp are constant the
output value is lower while the
time taken for the graph to reach
steady state is also shorter.

2 40 10 71.5 39.91 When the Kp value is higher and

τp is constant, both output value
and time taken for the graph to
reach steady state also increases.

3 10 20 120.5 9.96 Once the τp value increases and

Kp remains constant, the time
taken for the graph to reach
steady drastically increases
while the output value reduces.

4 30 5 39.5 29.92 When the Kp value is high and

the τp value is low,it takes less
time for the graph to reach
steady state.

5 20 10 67.5 19.94 When the τp value is about half

of the Kp value, it takes a longer
time to reach a steady state
compared to when the τp value is
smaller in a large range.

6 2 20 93.5 1.969 When the Kp value is reduced

drastically, it takes longer time
to reach a steady state when
compared to a large Kp value.

First-order systems are the simplest dynamic systems to analyze. A first order control system
is defined as a type of control system whose input-output relationship or also known as a
transfer function is a first-order differential equation[6]. During the first order system, process
gain, Kp and time constant, τp are the two major system inputs which affect the first order

6
system’s output. Each output profit is observed by adjusting the process gain and time constant
to analyse the dynamic response of a first order system. The graphs in the appendices clearly
indicate that a first order process does not respond to a step input simultaneously[2]. When the
Kp is kept constant, the time taken to reach steady state increases although the τp increases. But
when the τp increase the final steady state output value does not change. Moreover, when the
τp value increases and Kp remains constant, the time taken for the graph to reach steady
drastically increases while the output value reduces. However, an increase in Kp value, on the
other hand, reduces the time it takes for the process to reach a new steady state. The standard

form of first order of transfer function:

TASK B: STUDY OF SECOND ORDER SYSTEM

No. Kp A B Type Overshoot Decay Rise Settling Period

Ratio Time Time (s)
(s) (s)

1 10 40 14 Overdamped - - - - -

2 10 10 1 Underdamped 0.599 0.357 15.49 81.99 20

3 10 46 13.5 Critically - - - - -
damped

4 6 6.25 5 Critically - - - - -
damped

5 20 40 15 Overdamped - - - - -

6 20 5 10 Overdamped - - - - -

7
 𝐾𝑝 𝐾𝑝
The general second order transfer function: 𝐺(𝑠) = 𝜏2𝑝𝑆2 +2𝜁𝜏𝑝𝑆+1 = 𝐴𝑠2 +𝐵𝑠+1 (Seborg et al.,
2011). According to Seborg et al., 2011, 𝝃 > 1 is overdamped, 𝝃 = 1 is critically damped and
0 < 𝝃 < 1 is underdamped.

Case 1: Case 2: Case 3:

A = 𝜏𝑝2 = 40 A = 𝜏𝑝2 = 10 A = 𝜏𝑝2 = 45.5

𝜏𝑝 = 6.32 𝜏𝑝 = 3.16 𝜏𝑝 = 6.78
B = 2𝜁𝜏𝑝 = 14 B = 2𝜁𝜏𝑝 = 1 B = 2𝜁𝜏𝑝 = 13.5
𝜁 = 1.11 > 1 𝜁 = 0.16 < 1 𝜁=1
(Overdamped) (Underdamped) (Critically damped)

Case 4: Case 5: Case 6:

A = 𝜏𝑝2 = 6.25 A = 𝜏𝑝2 = 40 A = 𝜏𝑝2 = 5

𝜏𝑝 = 2.5 𝜏𝑝 = 6.32 𝜏𝑝 = 2.24
B = 2𝜁𝜏𝑝 = 5 B = 2𝜁𝜏𝑝 = 15 B = 2𝜁𝜏𝑝 = 10
𝜁=1 𝜁 = 1.19 < 1 𝜁 = 2.23 > 1
(Critically damped) (Overdamped) (Overdamped)

Underdamped system

𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑝𝑒𝑎𝑘 𝑜𝑢𝑡𝑝𝑢𝑡 𝑣𝑎𝑙𝑢𝑒−𝐹𝑖𝑛𝑎𝑙 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒 𝑎

Overshoot (OS) = =
𝐹𝑖𝑛𝑎𝑙 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒−𝐼𝑛𝑖𝑡𝑖𝑎𝑙 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒 𝑏

𝑆𝑒𝑐𝑜𝑛𝑑 𝑚𝑎𝑥𝑖𝑚𝑢𝑚 𝑝𝑒𝑎𝑘 𝑜𝑢𝑡𝑝𝑢𝑡 𝑣𝑎𝑙𝑢𝑒−𝐹𝑖𝑛𝑎𝑙 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒 𝑐

Decay Ratio (DR) = =𝑏
𝑀𝑎𝑥𝑖𝑚𝑢𝑚 𝑝𝑒𝑎𝑘 𝑜𝑢𝑡𝑝𝑢𝑡 𝑣𝑎𝑙𝑢𝑒− 𝐹𝑖𝑛𝑎𝑙 𝑠𝑡𝑒𝑎𝑑𝑦 𝑠𝑡𝑎𝑡𝑒 𝑣𝑎𝑙𝑢𝑒

Rise time (tr) is defined as the time taken for the process response to reach its final steady
state value for the first time.
𝜏
tr = (𝜋 − cos−1 𝜁), where 𝜁 radian mode.
√1− 𝜁 2

Settling time (ts) is defined as the time taken for the process response to reaches 95% final
steady state value.
Period (P) is the time between 2 consecutive peaks.

Case 2:
𝑎 15.99−10
OS = 𝑏 = =0.599
10−0
𝑐 12.14−10
DR = 𝑏 = =0.357
15.99−10

8
TASK C: DETERMINE PARAMETERS OF FIRST AND SECOND ORDERS SYSTEM

SYSTEM INDENTIFICATION PROBLEM 1:

Figure 3.1: The slope of initial response

3.2−0.9
Slope of initial response = (68−13)𝑠 = 0.04𝑠 −1

The final output value minus the initial output value

∆ Output value = Final output value – initial output value

= 3.2 – 0

= 3.2

9
The table with the parameter values calculated and the first order transfer function of
this unknown system

∆ 𝑂𝑢𝑡𝑝𝑢𝑡 3.2−0
KP = = = 0.2133
∆ 𝐼𝑛𝑝𝑢𝑡 15−0

After a time interval equal to the process time constant (t = τp), the first order process process
is still only 63.21% complete. (Seborg et al., 2011)

63.21 % ∆ output = 0.6321 x 3.2 = 2.023

Referring to Figure A-7,

τp = The time taken by the process to achieve 63.21% complete = 42 s

Hence, the First Order Transfer Function of unknown system is:

𝐾𝑝 0.2133
𝑌(𝑠) = 𝜏𝑝 𝑠 +1
= 42𝑠+1

10
SYSTEM INDENTIFICATION PROBLEM 2:

Δ Output value = Final output value – initial output value

= 14.79 – 0 = 14.79
∆ 𝑂𝑢𝑡𝑝𝑢𝑡 𝑣𝑎𝑙𝑢𝑒 14.79 − 0
𝐾𝑝 = = = 98.6
∆ 𝐼𝑛𝑝𝑢𝑡 𝑣𝑎𝑙𝑢𝑒 0.15 − 0

Table 3.1: Transfer function for Unknown System

Kp 98.6
𝜏p 4.5873
𝝃 0.2345

Thus, the Second order transfer function for this unknown system at step input of 0 to 0.15 is
𝐾𝑝 98.6 98.6
𝐺(𝑠) = = =
𝜏 2 𝑝𝑆 2 2 2 2
+ 2𝜁𝜏𝑝𝑆 + 1 4.5873 𝑠 + 2(0.2345)(4.5873)𝑠 + 1 21.0433𝑠 + 2.1514𝑠 + 1

11
4.0 DISCUSSION

4.1 Discuss the effect of gain increment/decrement on the system output.

Based on the results obtained in the first order system, the system output is higher when
the gain value is increased. The time taken for the system to reach steady state is also considered
short for higher gain value. As per the tabulation of data for task A, it can be seen that the
output obtained is highest which is 39.91 when the gain value is at highest where Kp= 40, while
comparing when the gain is the lowest where Kp = 2, the output value obtained was 1.969 only.
Thus, it can be concluded that the higher the gain value, the better the system performs with
larger output value and faster the system reaches steady state.

4.2 Discuss the effect of time constant increment/decrement on the system output.

The time constant greatly affects the time taken to reach the steady state rather than affecting
the system output. Based on the results, as the time constant, τp increases, the time taken to the
system reaches steady state increases while output value decreases. When the time constant is
the highest, τp = 20 with gain value, Kp =10, the output value decreases to 9.96 while the time
taken for the system to reach steady state is 120.5s which is the highest. However, while
comparing with constant gain Kp =10 and lower time constant which is τp = 10, the time taken
for the system to reach steady state decreases to 66s with the same system output value.

4.3 Explain on what happened to system output when negative value is used for gain.
Predict and illustrate the expected behaviour.

Based on the general first order transfer function shown below, Kp can be replaced with -Kp
to produce an inverse response transfer function model and the system output.

It is expected that the system output will decline when negative gain value is introduced to the
system. It indicates that when the gain is negative and a positive step input is used, the output
will move in the negative direction.

12
4.4 Explain on what happened to system output when negative value is used for time
constant. Predict and illustrate the expected behaviour.

Using a negative value for time constant cannot be done in system analysis and it is also
impossible to obtain output value. Time constant is defined as the time required for the output
to achieve 63.2 percent of its steady-state value. It is evident from this that it is intended for a
stable system. If a negative value for time constant is used, analytically the system output will
be unstable.

The exponential term involved in the time-domain response above illustrates the impossibility
of discovering system output with a negative time constant. A first order process also does not
react instantly to a sudden change in its input.

4.5 Evaluate the expected response from a first order and second order system driven
by a sinusoidal input?

The sinusoidal output of a system in terms of its magnitude, phase and noise, is given by: y(t)
= bsin (ωt + ϕ ) + n(t) where n is the noise, t is the time, b is the amplitude, ω is the angular
frequency and φ is the phase shift, related to the time shift by ϕ =ω⋅∆t.

Figure 4.1: Sinusoidal Input u(t) and Sinusoidal Response y(t)

13
For first order system, the standard form is given as below:

where Y (s) and U(s) are the laplace transforms of the output and input variables, respectively,
where K is the gain, and τ is the time constant. For a sinusoidal input,

The output signal is a sinusoid that has the same frequency, ω, as the input signal, u(t) =
Asin(ωt). The output has a phase shift, φ, relative to the input. The amount of phase shift
depends on ω. For a first order system, there are no peaks in gain. The frequency response
diagrams for a standard first-order system is shown below:

Figure 4.2: First order system response

For second order system, it is forced by a sinusoidal input, A sin 𝜔𝑡, the output for large
values of time which obtained after exponential terms as shown below:

In this case, the normalized amplitude ratio is independent of steady state gain[2]. For second
order response, the output has a different amplitude than the input. The response will be
oscillatory. The Figure 4.3 below shows the common second order response.

14
 Figure 4.3: Step response of the second-order system for selected damping ratios

4.6 In your opinion, explain on why second order has three different types of response.

The damping ratio describes how quickly the oscillations decay from one bounce to the next.
In most cases, oscillation is initiated when a system is interrupted from its initial position. There
are three damping ratios, indicated by ζ (zeta), which range from underdamped (ζ < 1) where
the poles are negative real with conjugate dynamics, critically damped (ζ = 1) where both poles
are in balance, real and negative and overdamped (ζ > 1) when both poles are real and negative.
The Figure 4.4 below shows the dynamic response of second order systems for underdamped,
critically damped and overdamped responses.

Figure 4.4: Second order step response

As per the previous statement, damping is an influence inside or on an oscillatory system that
reduces or prevents its oscillation. Control engineering, chemical engineering, mechanical
engineering, structural engineering, and electrical engineering are all interested in the

15
behaviour of oscillating systems. The physical quantity oscillating varies significantly and
might be the swaying of a tall building in the wind or the speed of an electric motor, but a
normalised, or non-dimensionalised approach can be useful in characterising common
elements of behaviour.

4.7 Explain the significance of studying first order and second order system based on
control perspective in chemical engineering.

For the purposes of analysis and control system design, physical systems can be modelled using
first- and second-order systems. Most real-world systems can be mathematically represented
as first-order systems. First order systems are distinguished by their ability to store material,
energy, or momentum, as well as the resistance associated with the flow of mass, energy, or
momentum that exceeds the capacity. Some examples include a change in the temperature in
an oven due to heat input and a change in the liquid level in a tank in chemical industries. The
first-order system does not exhibit overshoot or oscillatory behavior, making it unsuitable for
certain situations. A second-order model is more appropriate in such cases. Systems such as
these can exhibit a variety of responses to input stimuli using a second-order model, such as
overshoot, undershoot or oscillatory behavior. Examples include spring-mass-damper system
and car suspension system[5]. First-order and second-order systems are easy to analyze and
design. Hence they are widely used in preliminary analysis and design related to real-world
systems.

16
5.0 CONCLUSION AND RECOMMENDATION

In a nutshell, the purpose of this study, which is to identify the dynamics of the first and second
order systems, have been achieved. MATLAB is used to determine the dynamic responses of
the first and second order systems. Various input values were utilized in this experiment,
resulting in different effects on the results. For the first order system, process gain, Kp and
time constant, τp are the two system inputs which affects it’s output results. Moreover, higher
the gain value, the better the system performs with larger output value and faster the system
reaches steady state. Meanwhile, for the second order system there were three damping ratios
which was indicated by ζ (zeta). They were range from underdamped (ζ < 1), critically damped
(ζ = 1) as well as overdamped (ζ > 1). During underdamped response, it showed greater decay
ratio, overshoot and period. But during critical and overdamped processes, the overshoot, decay
ratio and period can be negligible. Thus, process gain determines the final result, time constant
determines whether the response is quick or slow, and zeta value determines the degree of
oscillation and damping in the system. As a recommendation, in MATLAB, the graphs that
were displaced was small at the same time it caused difficulty to read the results from the
graphs. In order to solve this problem, either we have to zoom in the graph until we can clearly
see the data.

17
REFERENCE

[1] Angeles, J. (2011). Time response of first- and second-order Dynamical Systems.
Mechanical Engineering Series, 85–231. https://doi.org/10.1007/978-1-4419-1027-1_2

[2] Seborg, D., Edgar, T., Mellichamp, D., & Doyle III, F. (2011). Process Dynamic and
Control. United State: Wiley.

[3] Second Order Systems - et.byu.edu. (n.d.). Retrieved April 28, 2022, from
https://www.et.byu.edu/~tom/classes/436/ClassNotes/Class20(Second-Order).pdf

[4] WatElectronics. (2022, February 3). Damping ratio in control system : Formula & Its
Significance. WatElectronics.com. Retrieved April 29, 2022, from
https://www.watelectronics.com/damping-ratio/

[5] Haidekker, M. A. (2013). Solving Differential Equations in the Laplace Domain. Linear
Feedback Controls, 27–56. https://doi.org/10.1016/b978-0-12-405875-0.00003-6

[6] Electrical4U. (2021, January 24). First Order Control System: What is it? (Rise & Settling
Time Formula). Electrical4U. Retrieved April 27, 2022, from
https://www.electrical4u.com/first-order-control-
system/#:~:text=The%20first%2Dorder%20control%20system%20tells%20us%20the%20spe
ed%20of,i.e%20is%20the%20transfer%20function.

18
APPENDIX

TASK A: FIRST ORDER SYSTEM

Figure 1: Kp=10, τp =10

Figure 2: Kp=40, τp =10

19
Figure 3: Kp=10, τp =20

Figure 4: Kp=30, τp =5

20
Figure 5: Kp=20, τp =10

Figure 6: Kp=2, τp =20

21
TASK B: SECOND ORDER SYSTEM

Figure 7: Kp=10, A =40 B=14

Figure 8: Kp=10, A =10 B=1

22
Figure 9: Kp=10, A=46 B=13.5

Figure 10: Kp=6, A=6.25 B=5

23
Figure 11: Kp=30, A=10 B=5

Figure12: Kp=20, A=5, B=10

24