Continuous control System 3 rd year Automation EEA Department

Course Information

Instructor: Dr. Abdelhamid Bounemeur

Contact:

• E-mail: [email protected]
• Office: 533 (each Thursday from 14h-16h)

Schedule:

• 2 hours: course
• 1.5 hour: exercises session
• 1.5 hour: Lab (MATLAB Simulation)

Entering test:

• No test is considered in this course

Student’s background: 1-Time domain analysis

2-General ideas about system representation

Course Target (Syllabus):

Generally, your course is divided in three chapters:

1. Why frequency domain.

2. System response to sinusoidal input signal.
3. Frequency response representation
4. Polar plot (Nyquist plot).
5. Stability study in polar plot.
 Continuous control System 3 rd year Automation EEA Department

6. Bode plot.
7. Stability study in Bode plot.
8. Nyquist stability criterion.
9. Black-Nichols Diagram.
10. Why avoiding PI, PD controller.
11. Design of feedback control systems (Lead and Lag Compensators), often called
correction.

Evaluation:

1. Continuous:
➢ 40% including:
• Interaction in both courses and exercises
• Labs.
• Interrogations
• Portfolio
2. Exams
➢ 60%Final exam including all chapters presented.
Continuous control System 3 rd year Automation EEA Department

Starting Ideas

Continuous control
system

Time-Domain Frequency
domain

Stability
assessment

Steady-state Stability
study assessment

Transient Revers
response Criterion
Steady-state Nyquist criterion
specification
s
Resonant
Gain margin
frequency

Bandwidth
Phase margin
Resonant
peak
Continuous control System 3 rd year Automation EEA Department

Chapter 01

Frequency Response (Graphical Representation)

Important Tips

• The course duration is about five weeks, with one exercise serie and one Lab.
• Course Support: “Katsuhiko Ogata” book.

1. Introduction

I
n this chapter, we consider the steady-state response of a system to a sinusoidal input test
signal. We will see that the response of a linear constant coefficient system (time-invariant) to
a sinusoidal input signal is an output sinusoidal signal at the same frequency as the input.
However, the magnitude and phase of the output signal differ from those of the input sinusoidal
signal, and the amount of difference is a function of the input frequency. Thus, we will be
investigating the steady-state response of the system to a sinusoidal input as the frequency varies.
We will examine the transfer function 𝐺(𝑠) when 𝑠 = 𝑗𝜔 and develop methods for graphically
displaying the complex number 𝐺(𝑗𝜔) as 𝜔 varies. We will start by considering polar plots and log-
magnitude and phase diagrams, and then the Bode plot. It’s well-known that Bode plot is one of the
most powerful graphical tools for analyzing and designing control systems, and we will cover that
subject in this chapter.

The outline of this chapter is summarized below:

• Frequency Response.
• frequency domain specifications for second-order system
• Polar plot
• Stability using Polar plot
• Bode Plot
Continuous control System 3 rd year Automation EEA Department

• Stability in Bode plot

• Nyquist Stability criterions
• Black-Nichols Diagram
• Stability using Black-Nichols Diagram
Continuous control System 3 rd year Automation EEA Department

1. Frequency Response

The frequency response of a system is defined as the steady-state response of the system to a
sinusoidal input signal. The sinusoid is a unique input signal, and the resulting output signal for a
linear system is sinusoidal in the steady state; it differs from the input only in amplitude and phase
angle.

Let consider the system block below:

u(t) System y(t)
g(t)
when, 𝑢(𝑡) = 𝐴𝑠𝑖𝑛(𝜔𝑡), we suppose to obtain in the output 𝑦(𝑡) = 𝐴́𝑠𝑖𝑛(𝜔𝑡 + 𝜙)

Ty to obtain the value of 𝐴́ and 𝜙 Proof (in the course)

Definition: The frequency response of a time-invariant linear system is given by the variation of
the magnitude 𝐴́ and the phase 𝜙, when the input frequency is changing. Thus, the steady-state
output signal depends only on the magnitude and phase of 𝐺(𝑗𝜔) at a specific frequency 𝜔.

One advantage of the frequency response method is the ready availability of sinusoid test signals
for various ranges of frequencies and amplitudes. Thus, the experimental determination of the
system frequency response is easily accomplished. The unknown transfer function of a system can
often be deduced from the experimentally determined frequency response of a system. Furthermore,
the design of a system in the frequency domain provides the designer with control of the bandwidth
of a system, as well as some measure of the response of the system to undesired noise and
disturbances.

A second advantage of the frequency response method is that the transfer function describing the
sinusoidal steady-state behavior of a system can be obtained by replacing 𝑠 with 𝑗𝜔 in the system
transfer function 𝐺(𝑗𝑤). The transfer function representing the sinusoidal steady-state behavior of
a system is then a function of the complex variable 𝑗𝜔 and is itself a complex function 𝐺(𝑗𝑤) that
possesses a magnitude and phase angle. The magnitude and phase angle of 𝐺(𝑗𝑤) are readily
Continuous control System 3 rd year Automation EEA Department

represented by graphical plots that provide significant insight into the analysis and design of control
systems.

The basic disadvantage of the frequency response method for analysis and design is the indirect
link between the frequency and the time domain. Direct correlations between the frequency
response and the corresponding transient response characteristics are somewhat tenuous, and in
practice the frequency response characteristic is adjusted by using various design criteria that will
normally result in a satisfactory transient response.

1.1.1- Magnitude and phase of second order system

The typical transfer function of the second-order system is given by the following relationship:

𝜔𝑛2
𝑇(𝑠) =
𝑠 2 + 2𝜁𝜔𝑛 𝑠 + 𝜔𝑛2

where 𝜔𝑛: Natural frequency (undamped)

𝜁: Damping coefficient

The first step in frequency response is to replace 𝑠 by 𝑗𝜔, then calculate the magnitude and the
phase (see the development in the course).

1.1.2- Resonant frequency 𝝎𝒓

It is the frequency at which the output magnitude is maximum.

Depending on this definition, try to find the mathematical formula of 𝝎𝒓 (see the development in
the course).

1.1.3- Resonant peak 𝑴𝒑

It is the peak value of the magnitude of the transfer function 𝑇 (𝑗𝜔).

Depending on this definition, try to find the mathematical formula of 𝑴𝒑 (see the development in
the course).
Continuous control System 3 rd year Automation EEA Department

1.1.4- Bandwidth

It is the range of frequencies over at which the magnitude of the frequency response is equal or
1
greater than −3𝑑𝐵 or or 70.7% from the zero initial value.
√2

Depending on this definition, try to find the mathematical formula of the bandwidth (see the
development in the course).

Example. 1.1

A second order system has a maximum overshoot of 50% and period of damped oscillation 0.25𝑠
in step response, determine:

• Resonant frequency
• Resonant peak
• Bandwidth

Solution (in the course)

Example. 1.2

For the second order system with unity feedback 𝐻(𝑠) = 1

200
𝐺( 𝑠 ) =
𝑠 (𝑠 + 8 )

• Find the frequency response specifications

Solution (in the course)

2. Frequency Response Plots

The transfer function of a system 𝐺(𝑠) can be described in the frequency domain by the relation
Continuous control System 3 rd year Automation EEA Department

Alternatively, the transfer function can be represented by a magnitude |𝐺(𝑗𝜔)| and phase as 𝜙(𝑗𝜔)

where

The graphical representation of the frequency response of the system 𝐺(𝑗𝜔) can utilize either the
first or the second representation given below.

2.1. Polar Plot

The polar plot of a sinusoidal transfer function 𝐺(𝑗𝜔) is a plot of the magnitude of 𝐺(𝑗𝜔) versus
the phase angle of 𝐺(𝑗𝜔) on polar coordinates as 𝜔 is varied from zero to infinity. Thus, the polar
plot is the locus of vectors |𝐺(𝑗𝜔)|∠𝐺(𝑗𝜔) as 𝜔 is varied from zero to infinity. Note that in polar
plots a positive (negative) phase angle is measured counterclockwise (clockwise) from the positive
real axis.

The polar plot is often called the Nyquist plot. Each point on the polar plot of 𝐺(𝑗𝜔) represents the
terminal point of a vector at a particular value of 𝜔. In the polar plot, it is important to show the
frequency graduation of the locus. The projections of 𝐺(𝑗𝜔) on the real and imaginary axes are its
real and imaginary components.

The polar plot representation of the frequency response is obtained by using Equations above. The
coordinates of the polar plot are the real and imaginary parts of 𝐺(𝑗𝜔), as shown in Figure. 1.
Continuous control System 3 rd year Automation EEA Department

Figure. 1

An example of a polar plot will illustrate this approach. The simplest steps to sketch a polar plot are
given below:
• Replace 𝑠 by 𝑗𝜔 in the transfer function 𝐺(𝑠)
• Calculate the real part and the imaginary part of 𝐺(𝑗𝜔) (if needed).
• Calculate the module and phase of 𝐺(𝑗𝜔).
• Replace 𝜔 from 0 𝑡𝑜 + ∞ and calculate the real part, the imaginary part, module and phase
of 𝐺(𝑗𝜔) (use a table).
Example. 1 First order factor.
𝐺(𝑠) = (1 + 𝜏𝑠)
where 𝜏 represent the time-constant of the system

• Sketch the polar plot

Example. 2 Integral factor.

1
𝐺( 𝑠 ) =
𝜏𝑠

where 𝜏 represent the time-constant of the system

• Sketch the polar plot

Example. 3 Derivative factor.

Continuous control System 3 rd year Automation EEA Department

𝐺(𝑠) = 𝜏𝑠

where 𝜏 represent the time-constant of the system

• Sketch the polar plot

Example. 4 (Frequency response of an RC filter) A simple RC filter is shown below.

• Find the transfer function of this filter.

• Sketch the polar plot following the steps below.

Example. 5 A simple second order transfer function is shown below.

𝑘
𝐺( 𝑠 ) =
(1 + 𝜏2 𝑠)(1 + 𝜏1𝑠)

• Sketch the polar plot

2.1.1. Advantages of Polar plot
• Polar plot explains the frequency response characteristic by only single plot.
• Polar plot study the stability easily comparing with Bode plot.
• Easy to obtain gain and phase margin.
• Polar plot need only the open-loop transfer function (OLTF) to find the frequency response.
2.1.2. Disadvantages of Polar plot (drawbacks)
• The polar plot doesn’t clearly indicate the contribution of each individual factor of the open-
loop transfer function.
Continuous control System 3 rd year Automation EEA Department

Remark: Never forget that Polar plot is also called Nyquist plot

3. Stability using Polar plot

Stability is the essence of control system analysis. Automation engineer must know almost all the
tools to determine whether the system is stable or no. during the time-domain analysis studied in
the last chapters, we have seen that all poles of the closed-loop system must lie in the left half side
(LHS), and due the hardness to calculate the poles when the degree of the characteristic equation is
higher. Routh criterion was used to determine the stability of the system without calculating the
poles. Moreover, Routh criterion can determine the number of the poles lie in the right half side
(RHS) without clearly indicates their values. In this chapter, graphical methods will be illustrated
in order to assess the stability, and the first one will be the Reverse criterion.

• Revers criterion: The criterion of Revers is a graphical tool because it relies on the graphical
representation of the open-loop transfer function as 𝜔 varies. This criterion is applicable only
to a system whose open-loop transfer function does not have poles and zeros with positive
real parts, this is also called a minimum phase system.
To study the stability using this criterion, we should check the enclosement around the critical
point (-1,0), if the polar plot encloses the critical point, so the system is said to be unstable,
if not, the system is stable. In another words, if the critical point is at the left of the polar plot
when 𝜔 is increasing, so the system is stable, otherwise the system is unstable.
Why the critical point is chosen (-1,0), this question is your first exercise in the first
home work.
3.1- Gain Margin and Phase Margin
• Gain Margin: It is defined as the factor by which the system gain can be increased to drive
the system to the verge of instability. In other word, the gain margin represents the additional
gain a system can tolerate before it reaches instability. This margin is computed by
determining the gain at the phase crossover frequency.
Continuous control System 3 rd year Automation EEA Department

• Phase crossover frequency: It is the frequency at which the phase angle of the open -
loop transfer function equals –180°.
The plot of the gain margin is given below

The formula for gain margin in a polar plot is typically given as:
1
Gain Margin 𝐺𝑎𝑖𝑛 𝑀𝑎𝑟𝑔𝑖𝑛 (𝐺𝑀) = |
𝐺 (𝑗𝜔 𝑝𝑐)|

This gain margin can be written in terms of decibel:

1
𝐺𝑎𝑖𝑛 𝑀𝑎𝑟𝑔𝑖𝑛 (𝐺𝑀)𝑑𝑏 = 20𝑙𝑜𝑔 ( ) = −20𝑙𝑜𝑔(|𝐺(𝑗𝜔𝑝𝑐 )|)
|𝐺(𝑗𝜔𝑝𝑐 )|
where
• 𝜔𝑝𝑐 is the phase crossover frequency (frequency at which the phase angle is -180 degrees).
• |𝐺(𝑗𝜔𝑝𝑐 )| is the gain of the system at the phase crossover frequency.
This formula represents the reciprocal of the magnitude of the system's gain at the phase
crossover frequency. It essentially quantifies the amount of additional gain the system can
handle at that frequency before it becomes unstable.
Continuous control System 3 rd year Automation EEA Department

If the gain margin in decibel is positive, then the system is stable, and if it is negative, so the
system is said to be unstable.
• Phase Margin: The phase margin in a polar plot refers to the additional phase shift a system
can tolerate before it reaches instability. It is calculated as the difference between the phase
angle of the system at the frequency where the gain is unity (0 dB) and the phase angle at the
gain crossover frequency where the magnitude of the system's transfer function is unity.
• Gain cross over frequency: It is the frequency at which the magnitude of the open loop
transfer function is unity.
The plot of the phase margin can be given as follow:

The formula for phase margin in a polar plot is:

𝑝ℎ𝑎𝑠𝑒 𝑀𝑎𝑟𝑔𝑖𝑛 (𝑃𝑀) = 180° + 𝑝ℎ𝑎𝑠𝑒(𝐺(𝑗𝜔𝑔𝑐 ))
where
• 𝜔𝑔𝑐 : is the gain crossover frequency (frequency at which the magnitude of the system's
transfer function is unity).
• 𝑝ℎ𝑎𝑠𝑒(𝐺(𝑗𝜔𝑔𝑐 ): represents the phase angle of the system at the gain crossover frequency.
Continuous control System 3 rd year Automation EEA Department

In a polar plot, the phase margin can be determined by finding the phase angle at the gain
crossover frequency (𝜔𝑔𝑐 ), this result represents the phase margin and indicates how much
additional phase shift the system can handle before it becomes unstable.
Examples

Consider the transfer function 𝐺(𝑠) placed in unity feedback system

5
𝐺 (𝑠 ) =
𝑠 3
(1 + )
100
• Sketch the Nyquist plot (polar plot).
• Assess the system is whether stable or unstable using Revers criterion.
• Define the phase crossover frequency.
• Define the gain crossover frequency
• Define the gain margin and the phase margin

4. Bode plot
The introduction of logarithmic plots, often called Bode plots, simplifies the determination of
the graphical portrayal of the frequency response. The logarithmic plots are called Bode plots
in honor of H. W. Bode, who used them extensively in his studies of feedback amplifiers. The
transfer function in the frequency domain is

Bode plot is applicable only if the system is minimum-phase system (all poles and zeros lie in the
left-half side). The Bode plot or the Bode diagram consists of two plots:
• Magnitude plot
• Phase plot
In both of the plots, x-axis represents angular frequency (logarithmic scale). Whereas, y-axis
represents the magnitude (linear scale) of open loop transfer function in the magnitude plot and the
phase angle (linear scale) of the open loop transfer function in the phase plot.
Continuous control System 3 rd year Automation EEA Department

The magnitude of the open loop transfer function in dB is

The phase angle of the open-loop transfer function in degrees is

Note − The base of logarithm is 10.

4.1- Bode plot of basic functions:
• Constant term: For the gain K, the magnitude is 𝟐𝟎 𝒍𝒐𝒈𝟏𝟎 𝑲 and the phase is 0°; both are
constant with frequency. Thus, the magnitude and phase plots of the gain are shown in the
Figure (a: gain, b: phase) below
Continuous control System 3 rd year Automation EEA Department

• Pole/zero at the origin: For the zero (𝒋𝝎) at the origin, the magnitude is 20 𝑙𝑜𝑔10 𝜔 and the
phase is 90° . These are plotted in the following Figure (a: gain, b: phase), where we notice
that the slope of the magnitude plot is 20 𝑑𝐵/𝑑𝑒𝑐𝑎𝑑𝑒, while the phase is constant with
1
frequency. The corner frequency is 𝑧 = 1 𝑟𝑎𝑑/𝑠 (generally, the corner frequency 𝑧 = ,
𝑇

𝑇:time constant).

Remark: A decade is an interval between two frequencies with a ratio of 10; e.g., between 𝜔0 and
10𝜔0 , or between 10 and 100 𝐻𝑧. Thus, 20 𝑑𝐵/𝑑𝑒𝑐𝑎𝑑𝑒 means that the magnitude changes 20 𝑑𝐵
whenever the frequency changes tenfold or one decade.
𝟏
The Bode plots for the pole are similar except that the slope of the magnitude plot is −20
𝒋𝝎

dB/decade while the phase is −90°.

In general, for 𝒋𝝎𝑵 , where 𝑁 is an integer, the magnitude plot will have a slope of 20𝑁 𝑑𝐵/𝑑𝑒𝑐𝑎𝑑𝑒,
while the phase is 𝑁 ∗ 90°.
Continuous control System 3 rd year Automation EEA Department

𝟏
The same thing when we have where 𝑁 is an integer, the magnitude plot will have a slope of
𝒋𝝎𝑵

−20𝑁 𝑑𝐵/𝑑𝑒𝑐𝑎𝑑𝑒, while the phase is −𝑁 ∗ 90°.

• Simple pole/zero: For the simple zero (𝟏 + 𝒋𝝎/𝒛𝟏 ), the magnitude is 20 𝑙𝑜𝑔10 |1 +
𝑗𝜔/𝑧1 | and the phase is 𝑡𝑎𝑛−1 (𝜔/𝑧1 ). We notice that

showing that we can approximate the magnitude as zero (a straight line with zero slope) for
small values of ω and by a straight line with slope 20 dB/decade for large values of ω.
The frequency 𝜔 = 𝑧1 where the two asymptotic lines meet is called the corner frequency
or break frequency (see the figure below, where a: gain plot and b: phase plot).
Continuous control System 3 rd year Automation EEA Department

Notice that the approximate plot is close to the actual plot except at the break frequency, where
𝜔 = 𝑧1 and the deviation is 20 𝑙𝑜𝑔10|(1 + 𝑗1)| = 20 𝑙𝑜𝑔10√2 = 3𝑑𝐵.
The phase 𝑡𝑎𝑛−1 (𝜔⁄𝑧1) can be expressed as

To make things easy, the phase plot is having phase angle of 0 degrees up to 𝝎 = 𝒛𝟏 𝒓𝒂𝒅/𝒔𝒆𝒄
and from here, it is having phase angle of 𝟗𝟎° .

4.2- Procedure to sketch the bode plot:

• Step1: write the open-loop transfer function in a standard form (with time constant).
• Step 2: Identify the corner frequencies of the transfer function in ascending order and give
the slope for each line.
• Step 3: calculate the resulting slope
• Step4: Identify the slope of the 1 st line for Bode plot (the first corner frequency) and shift the
graph depending on the static gain 𝐾.
• Step 5: Write the phase equation.
• Step 6: sketch the gain and deduce the phase plot
Example. 1: Construct the Bode plots for the transfer function
𝟐𝟎𝟎𝒔
(𝒔 + 𝟏𝟎)(𝟏𝟎𝟎 + 𝒔)

Example. 2: Construct the Bode plots for the transfer function

𝟏𝟒𝟒𝟎𝟎 (𝒔 + 𝟓)
𝒔𝟐 (𝒔 + 𝟐𝟎)(𝟏𝟎𝟎 + 𝒔)

Table of usual bode plot for basics transfer function: (in the course).
Continuous control System 3 rd year Automation EEA Department

Try to sketch the Bode plot for a second order system transfer function in two cases: 𝑮(𝒔) =
𝟏
𝒔𝟐 + 𝟐𝜻𝝎𝒏 𝒔 + 𝝎𝒏 𝟐 and 𝑮(𝒔) = (The first exercise of the first homework)
𝒔 𝟐 +𝟐𝜻𝝎𝒏 𝒔+𝝎𝒏 𝟐