1.control Systems Notes
1.control Systems Notes
NUST PNEC
ME 314
1 INTRODUCTION 2
1
1 INTRODUCTION
Control systems are an integral part of modern society. Control system Engineering is concerned with
understanding and controlling the materials and forces of nature for the benefit of humankind.
Control system engineers are concerned with understanding and controlling segments of their
environment, often called systems, to provide useful economic products for society.
The present challenge is the modelling and control of modern, complex, interrelated systems such as
traffic control systems, chemical processes, and robotic systems.
A control system consists of subsystems and processes (or plants) assembled for
the purpose of obtaining a desired output with desired performance, given a
specified input.
1. Power Amplication
2. Remote Control
3. Convinience of input form
4. Compensation for disturbances
2
1.2 OPEN – LOOP CONTROL SYSTEM
A generic open-loop system is shown in Figure 1-2. It starts with a subsystem called an input
transducer, which converts the form of the input to that used by the controller. The controller drives
a process or a plant. The input is sometimes called the reference, while the output can be called the
controlled variable. Other signals, such as disturbances, are shown added to the controller and process
outputs via summing junctions, which yield the algebraic sum of their input signals using associated
signs.
The distinguishing characteristic of an open-loop system is that it cannot compensate for any
disturbances that add to the controller's driving signal
The input transducer converts the form of the input to the form used by the controller. An output
transducer, or sensor, measures the output response and converts it into the form used by the
controller.
The closed-loop system compensates for disturbances by measuring the output response, feeding that
measurement back through a feedback path, and comparing that response to the input at the
summing junction. If there is any difference between the two responses, the system varies
instructions, via the actuating signal, to make a correction. If there is no difference, the system does
not change, since the system’s response is already the desired response.
3
A closed-loop control system uses a measurement of the output and feedback
of this signal to compare it with the desired output (reference or command).
4
2 MODELLING IN THE FREQUENCY DOMAIN
To understand and control complex systems, one must obtain quantitative mathematical models of
these systems. It is necessary therefore to analyse the relationships between the system variables and
to obtain a mathematical model. Because the systems under consideration are dynamic in nature, the
descriptive equations are usually differential equations. Furthermore, if these equations can be
linearized, then the Laplace transform can be used to simplify the method of solution.
In practice, the complexity of systems and the ignorance of all the relevant factors necessitate the
introduction of assumptions concerning the system operation. Therefore, it is often useful to consider
the physical system, express any necessary assumptions, and linearize the system. Then, by using the
physical laws describing the linear equivalent system, obtain a set of linear differential equations.
Finally, using mathematical tools, such as the Laplace transform, obtain a solution describing the
operation of the system. In summary, the approach to dynamic system modelling can be listed as
follows:
The usual system analysis workflow involves the use of Laplace transform function (discussed later)
on the mathematical model of the system. These models should be continuous for Laplace transform
to be applicable. Following types of signals are usually encountered during system analysis:
5
1. Obtain the linearized differential equations.
2. Obtain the Laplace transformation of the differential equations.
3. Solve the resulting algebraic equation for the transform of the variable of interest.
Where
1 𝜎𝜎+𝑗𝑗∞
ℒ −1 [𝐹𝐹(𝑠𝑠)] = � 𝐹𝐹(𝑠𝑠)𝑒𝑒 𝑠𝑠𝑠𝑠 𝑑𝑑𝑑𝑑 = 𝑓𝑓(𝑡𝑡)𝑢𝑢(𝑡𝑡)
2𝜋𝜋𝜋𝜋 𝜎𝜎−𝑗𝑗∞
• ℒ [𝛿𝛿(𝑡𝑡)] = 1
• ℒ [𝑢𝑢(𝑡𝑡)] = 1/𝑠𝑠
• ℒ [𝐴𝐴𝐴𝐴(𝑡𝑡)] = 𝐴𝐴/𝑠𝑠
Similarly, a table (available as a separate document) can be generated to represent common forms of
equations that are encountered during analysis, thereby reducing the need to apply the above formula
for every calculation.
SOLUTION: Since the time function does not contain an impulse function, we
can replace the lower limit of the integral with 0
∞ ∞ ∞
𝐹𝐹(𝑠𝑠) = � 𝑓𝑓(𝑡𝑡)𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑑𝑑𝑑𝑑 = � 𝐴𝐴𝑒𝑒 −𝑎𝑎𝑎𝑎 𝑒𝑒 −𝑠𝑠𝑠𝑠 𝑑𝑑𝑑𝑑 = 𝐴𝐴 � 𝑒𝑒 −(𝑠𝑠+𝑎𝑎)𝑡𝑡 𝑑𝑑𝑑𝑑
0 0 0
∞
𝐴𝐴 𝐴𝐴
=− 𝑒𝑒 −(𝑠𝑠+𝑎𝑎)𝑡𝑡 � =
𝑠𝑠 + 𝑎𝑎 𝑡𝑡=0 𝑠𝑠 + 𝑎𝑎
6
2.2 PROPERTIES OF LAPLACE TRANSFORMS
2.2.1 Linearity Theorem
If 𝐹𝐹1 (𝑠𝑠) and 𝐹𝐹2 (𝑠𝑠) are Laplacian transform of 𝑓𝑓1 (𝑡𝑡) and 𝑓𝑓2 (𝑡𝑡) respectively and 𝛼𝛼1 and 𝛼𝛼2 are arbitrary
constants, then:
ℒ [𝛼𝛼1 𝑓𝑓1 (𝑡𝑡) + 𝛼𝛼2 𝑓𝑓2 (𝑡𝑡)] = 𝛼𝛼1 𝐹𝐹1 (𝑠𝑠) + 𝛼𝛼2 𝐹𝐹2 (𝑠𝑠)
• Additivity: If 𝑓𝑓(𝑡𝑡1 ) = 𝐹𝐹1 and 𝑓𝑓(𝑡𝑡2 ) = 𝐹𝐹2 , then 𝑓𝑓(𝑡𝑡1 + 𝑡𝑡2 ) = 𝐹𝐹1 + 𝐹𝐹2
• Homogeneity: If 𝑓𝑓(𝑡𝑡) = 𝐹𝐹, then 𝑓𝑓(𝛼𝛼𝛼𝛼) = 𝛼𝛼𝛼𝛼
• Shift Invariance: If 𝑓𝑓(𝑡𝑡) = 𝐹𝐹(𝑡𝑡), then 𝑓𝑓(𝑡𝑡 + 𝑠𝑠) = 𝐹𝐹(𝑡𝑡 + 𝑠𝑠)
If only first two conditions are met, then the system is called linear system. If the third condition is
also met, then the system is known as linear time invariance system.
ℒ�𝑓𝑓 (𝑛𝑛) (𝑡𝑡)� = 𝑆𝑆 𝑛𝑛 𝐹𝐹(𝑠𝑠) − 𝑠𝑠 𝑛𝑛−1 𝑓𝑓(0+ ) − 𝑠𝑠 𝑛𝑛−2 𝑓𝑓 ′ (0+ ) − ⋯ 𝑠𝑠𝑓𝑓 (𝑛𝑛−2) (0+ ) − 𝑓𝑓 (𝑛𝑛−1) (0+ )
𝐹𝐹(𝑠𝑠) 𝑓𝑓 −1 (0+ )
ℒ �� 𝑓𝑓(𝑡𝑡)� = +
𝑠𝑠 𝑠𝑠
7
2.2.8 Final Value Theorem
If 𝐹𝐹(𝑠𝑠) is the Laplace transform of 𝑓𝑓(𝑡𝑡), then
𝑁𝑁(𝑠𝑠) 𝑁𝑁(𝑠𝑠)
𝐹𝐹(𝑠𝑠) = =
𝐷𝐷(𝑠𝑠) (𝑠𝑠 + 𝑝𝑝1 )(𝑠𝑠 + 𝑝𝑝2 )(𝑠𝑠 + 𝑝𝑝3 ) … (𝑠𝑠 + 𝑝𝑝𝑛𝑛 )
𝐾𝐾1 𝐾𝐾2 𝐾𝐾3 𝐾𝐾𝑛𝑛
= + + …+
𝑠𝑠 + 𝑝𝑝1 𝑠𝑠 + 𝑝𝑝2 𝑠𝑠 + 𝑝𝑝3 𝑠𝑠 + 𝑝𝑝𝑛𝑛
𝑁𝑁(𝑠𝑠) ≡ 𝐾𝐾1 (𝑠𝑠 + 𝑝𝑝2 )(𝑠𝑠 + 𝑝𝑝3 ) … (𝑠𝑠 + 𝑝𝑝𝑛𝑛 ) + 𝐾𝐾2 (𝑠𝑠 + 𝑝𝑝1 )(𝑠𝑠 + 𝑝𝑝3 ) … (𝑠𝑠 + 𝑝𝑝𝑛𝑛 ) + ⋯
The values of 𝐾𝐾1 , 𝐾𝐾2 , 𝐾𝐾3 , … 𝐾𝐾𝑛𝑛 can be found by substituting the roots of denominator in the above
equation.
SOLUTION: Substitute the corresponding 𝐹𝐹(𝑠𝑠) for each term in the above
differential equation, and taking initial conditions to be zero, the Laplace
transform equation is
32
𝑠𝑠 2 𝑌𝑌(𝑠𝑠) + 12𝑠𝑠𝑠𝑠(𝑠𝑠) + 32𝑌𝑌(𝑠𝑠) =
𝑠𝑠
Solving for 𝑌𝑌(𝑠𝑠) yields
32 32
𝑌𝑌(𝑠𝑠) = 2
≡
𝑠𝑠(𝑠𝑠 + 12𝑠𝑠 + 32) 𝑠𝑠(𝑠𝑠 + 4)(𝑠𝑠 + 8)
Performing partial-fraction expansion of the right-hand term gives
32 𝐾𝐾1 𝐾𝐾2 𝐾𝐾3
𝑌𝑌(𝑠𝑠) = ≡ + +
𝑠𝑠(𝑠𝑠 + 4)(𝑠𝑠 + 8) 𝑠𝑠 𝑠𝑠 + 4 𝑠𝑠 + 8
Taking LCM and cancelling denominator on each side gives
32 ≡ 𝐾𝐾1 (𝑠𝑠 + 4)(𝑠𝑠 + 8) + 𝐾𝐾2 (𝑠𝑠)(𝑠𝑠 + 8) + 𝐾𝐾3 (𝑠𝑠)(𝑠𝑠 + 4)
Substituting roots of the denominator yields
• When 𝑠𝑠 = 0, 𝐾𝐾1 = 1
8
• When 𝑠𝑠 = −4, 𝐾𝐾2 = −2
• When 𝑠𝑠 = −8, 𝐾𝐾3 = 1
Hence
1 2 1
𝑌𝑌(𝑠𝑠) = − +
𝑠𝑠 𝑠𝑠 + 4 𝑠𝑠 + 8
Since each of the three component parts of the equation above is represented
as an 𝐹𝐹(𝑠𝑠) in Laplace transform tables, 𝑦𝑦(𝑡𝑡) is the sum of the inverse Laplace
transforms of each term. Hence,
𝑦𝑦(𝑡𝑡) = (1 − 2𝑒𝑒 −4𝑡𝑡 + 𝑒𝑒 −8𝑡𝑡 )𝑢𝑢(𝑡𝑡)
𝑁𝑁(𝑠𝑠) 𝑁𝑁(𝑠𝑠)
𝐹𝐹(𝑠𝑠) = =
𝐷𝐷(𝑠𝑠) (𝑠𝑠 + 𝑝𝑝1 ) (𝑠𝑠 + 𝑝𝑝2 )(𝑠𝑠 + 𝑝𝑝3 ) … (𝑠𝑠 + 𝑝𝑝𝑛𝑛 )
𝑟𝑟
To evaluate the values of 𝐾𝐾1 , 𝐾𝐾2 , … 𝐾𝐾𝑛𝑛 (numerators of the partial fractions) LCM is taken, denominator
is cancelled out from both sides and then the coefficients of ‘𝑠𝑠’ in expanded RHS is compared with LHS
(i.e. 𝑁𝑁(𝑠𝑠), the Numerator)
9
Hence
5 2.5 2.5
𝐹𝐹(𝑠𝑠) = + −
(𝑠𝑠 + 1)2 𝑠𝑠 + 1 𝑠𝑠 + 3
To evaluate 𝑓𝑓(𝑡𝑡), inverse Laplace transform has to be applied
5 2.5 2.5
𝑓𝑓(𝑡𝑡) = ℒ −1 � � + ℒ −1 � � − ℒ −1 � �
(𝑠𝑠 + 1) 2 𝑠𝑠 + 1 𝑠𝑠 + 3
Applying frequency shift theorem on the first term:
If
5
ℒ −1 [𝐹𝐹1 (𝑠𝑠)] = 𝑓𝑓1 (𝑡𝑡) = ℒ −1 � �
(𝑠𝑠 + 1)2
then
5 5
ℒ −1 [𝐹𝐹1 (𝑠𝑠 − 1)] = 𝑒𝑒 𝑡𝑡 𝑓𝑓1 (𝑡𝑡) = ℒ −1 � � = ℒ −1 � 2 �
(𝑠𝑠 − 1 + 1)2 𝑠𝑠
5
𝑒𝑒 𝑡𝑡 𝑓𝑓(𝑡𝑡) = ℒ −1 � 2 �
𝑠𝑠
5
𝑓𝑓(𝑡𝑡) = 𝑒𝑒 −𝑡𝑡 ℒ −1 � 2 �
𝑠𝑠
Now the equation becomes:
1 1 1
𝑓𝑓(𝑡𝑡) = 5𝑒𝑒 −𝑡𝑡 ℒ −1 � 2 � + 2.5ℒ −1 � � − 2.5ℒ −1 � �
𝑠𝑠 𝑠𝑠 + 1 𝑠𝑠 + 3
Since each of the three component parts of the equation above is represented
as an 𝐹𝐹(𝑠𝑠) in Laplace transform tables, 𝑓𝑓(𝑡𝑡) is the sum of the inverse Laplace
transforms of each term. Hence,
𝑓𝑓(𝑡𝑡) = (5𝑒𝑒 −𝑡𝑡 ⋅ 𝑡𝑡 + 2.5𝑒𝑒 −𝑡𝑡 − 2.5𝑒𝑒 −3𝑡𝑡 )𝑢𝑢(𝑡𝑡)
= (5𝑡𝑡 ⋅ 𝑒𝑒 −𝑡𝑡 + 2.5𝑒𝑒 −𝑡𝑡 − 2.5𝑒𝑒 −3𝑡𝑡 )𝑢𝑢(𝑡𝑡)
𝑁𝑁(𝑠𝑠) 𝑁𝑁(𝑠𝑠)
𝐹𝐹(𝑠𝑠) = =
𝐷𝐷(𝑠𝑠) (𝑠𝑠 + 𝑝𝑝1 )(𝑠𝑠 2 + 𝑎𝑎𝑎𝑎 + 𝑏𝑏) …
𝐾𝐾1 𝐾𝐾2 𝑠𝑠 + 𝐾𝐾3
= + 2 +⋯
𝑠𝑠 + 𝑝𝑝1 𝑠𝑠 + 𝑎𝑎𝑎𝑎 + 𝑏𝑏
Similar to the previous case, to obtain values for numerator of partial fractions (𝐾𝐾1 , 𝐾𝐾2 , 𝐾𝐾3 , …), LCM is
taken, then the numerator is expanded and coefficients of different powers of ‘𝑠𝑠’ in the expanded
equation are compared with the numerator.
10
EXAMPLE 2.4 LAPLACE TRANSFORM SOLUTION OF A DIFFERENTIAL EQUATION
PROBLEM: Given the following differential equation, solve for 𝑦𝑦(𝑡𝑡) if all initial
conditions are zero. Use the Laplace transform.
𝑑𝑑 2 𝑦𝑦 𝑑𝑑𝑑𝑑
2
+2 + 2𝑦𝑦 = 4𝑢𝑢(𝑡𝑡)
𝑑𝑑𝑡𝑡 𝑑𝑑𝑑𝑑
SOLUTION: Substitute the corresponding 𝐹𝐹(𝑠𝑠) for each term in the above
differential equation, and taking initial conditions to be zero, the Laplace
transform equation is
4
𝑠𝑠 2 𝑌𝑌(𝑠𝑠) + 2𝑠𝑠𝑠𝑠(𝑠𝑠) + 2𝑌𝑌(𝑠𝑠) =
𝑠𝑠
Solving for 𝑌𝑌(𝑠𝑠) yields
4
𝑌𝑌(𝑠𝑠) =
𝑠𝑠(𝑠𝑠 2 + 2𝑠𝑠 + 2)
2
Since 𝑠𝑠 + 2𝑠𝑠 + 2 has complex roots, the expression cannot be factorised any
more.
Performing partial-fraction expansion of the right-hand term gives
𝐾𝐾1 𝐾𝐾2 𝑠𝑠 + 𝐾𝐾3
𝑌𝑌(𝑠𝑠) = + 2
𝑠𝑠 𝑠𝑠 + 2𝑠𝑠 + 2
Taking LCM and cancelling denominator on each side, and expanding gives
4 = 𝐾𝐾1 (𝑠𝑠 2 + 2𝑠𝑠 + 2) + 𝑠𝑠(𝐾𝐾2 𝑠𝑠 + 𝐾𝐾3 )
4 = 𝐾𝐾1 𝑠𝑠 2 + 2𝐾𝐾1 𝑠𝑠 + 2𝐾𝐾1 + 𝐾𝐾2 𝑠𝑠 2 + 𝐾𝐾3 𝑠𝑠
4 = (𝐾𝐾1 + 𝐾𝐾2 )𝑠𝑠 2 + (2𝐾𝐾1 + 𝐾𝐾3 )𝑠𝑠 + (2𝐾𝐾1 )
Comparing the coefficients of ‘𝑠𝑠’ gives the following system of linear equations:
𝐾𝐾1 + 𝐾𝐾2 = 0
2𝐾𝐾1 + 𝐾𝐾3 = 0
2𝐾𝐾1 = 4
Solving the above system yields
• 𝐾𝐾1 = 2
• 𝐾𝐾2 = −2
• 𝐾𝐾3 = −4
Hence
2 2𝑠𝑠 + 4
𝑌𝑌(𝑠𝑠) = − 2
𝑠𝑠 𝑠𝑠 + 2𝑠𝑠 + 2
1 𝑠𝑠 + 2
= 2� − 2 �
𝑠𝑠 𝑠𝑠 + 2𝑠𝑠 + 2
1 𝑠𝑠 + 2
= 2� − 2 �
𝑠𝑠 𝑠𝑠 + 2𝑠𝑠 + 12 + 12
1 𝑠𝑠 + 1 + 1
= 2� − �
𝑠𝑠 (𝑠𝑠 + 1)2 + 12
1 𝑠𝑠 + 1 1
= 2� − − �
𝑠𝑠 (𝑠𝑠 2
+ 1) + 1 2 (𝑠𝑠 + 1)2 + 12
Since each of the three component parts of the equation above is represented
as an 𝐹𝐹(𝑠𝑠) in Laplace transform tables, 𝑦𝑦(𝑡𝑡) is the sum of the inverse Laplace
transforms of each term. Hence,
11
1 𝑠𝑠 + 1 1
𝑦𝑦(𝑡𝑡) = 2ℒ −1 � − − �
𝑠𝑠 (𝑠𝑠 2
+ 1) + 1 2 (𝑠𝑠 + 1)2 + 12
= 2[1 − 𝑒𝑒 −𝑡𝑡 cos 𝑡𝑡 − 𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡]
= 2 − 2𝑒𝑒 −𝑡𝑡 cos 𝑡𝑡 − 2𝑒𝑒 −𝑡𝑡 sin 𝑡𝑡
𝑅𝑅
𝑖𝑖(𝑡𝑡)
𝑣𝑣(𝑡𝑡) 𝐶𝐶
12
1 1
= 𝐼𝐼(𝑠𝑠) �𝑅𝑅 + �
𝑠𝑠 𝐶𝐶𝐶𝐶
𝑅𝑅𝑅𝑅𝑅𝑅 + 1
= 𝐼𝐼(𝑠𝑠) � �
𝐶𝐶𝐶𝐶
𝐶𝐶
𝐼𝐼(𝑠𝑠) =
𝑅𝑅𝑅𝑅𝑅𝑅 + 1
1
=
1
𝑅𝑅 �𝑠𝑠 + �
𝑅𝑅𝑅𝑅
1 1
= ×
𝑅𝑅 𝑠𝑠 + 1
𝑅𝑅𝑅𝑅
𝑖𝑖(𝑡𝑡) = ℒ −1 𝐼𝐼(𝑠𝑠)
1 1
= ℒ −1 � × �
𝑅𝑅 𝑠𝑠 + 1
𝑅𝑅𝑅𝑅
1 −1 1
=ℒ � �
𝑅𝑅 1
𝑠𝑠 +
𝑅𝑅𝑅𝑅
1 −� 1 �𝑡𝑡
= 𝑒𝑒 𝑅𝑅𝑅𝑅
𝑅𝑅
Substituting the values of resistance and capacitance
1 −𝑡𝑡� 1
�
𝑖𝑖(𝑡𝑡) = 6 𝑒𝑒 106 ×10−6
10
= 10−6 𝑒𝑒 −𝑡𝑡
When time is 𝑡𝑡 = 0, then
𝑖𝑖(𝑡𝑡) = 10−6
ℒ(Output)
𝑇𝑇. 𝐹𝐹 =
ℒ(Input)
13
EXAMPLE 2.6 TRANSFER FUNCTION OF A RC NETWORK
PROBLEM: Evaluate an expression for a transfer function describing the system
illustrated in the circuit below:
𝑅𝑅
① 𝐶𝐶 ② 𝑣𝑣o (𝑡𝑡)
𝑣𝑣𝑖𝑖 (𝑡𝑡)
1
𝑣𝑣𝑖𝑖 (𝑡𝑡) = 𝑖𝑖(𝑡𝑡)𝑅𝑅 + � 𝑖𝑖(𝑡𝑡) 𝑑𝑑𝑑𝑑 Loop ①
𝐶𝐶
1
𝑣𝑣o (𝑡𝑡) = � 𝑖𝑖(𝑡𝑡) 𝑑𝑑𝑑𝑑 Loop ②
𝐶𝐶
𝐼𝐼(𝑠𝑠) 𝐼𝐼(𝑠𝑠)
= 𝐶𝐶𝐶𝐶 ÷ 𝐶𝐶𝐶𝐶
𝐼𝐼(𝑠𝑠) 𝐼𝐼(𝑠𝑠)
𝐼𝐼(𝑠𝑠)𝑅𝑅 +
𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶
1
=
𝑅𝑅𝑅𝑅𝑅𝑅 + 1
1
=
𝜏𝜏𝜏𝜏 + 1
14
EXAMPLE 2.7 TRANSFER FUNCTION OF RLC NETWORK
PROBLEM: Evaluate an expression for a transfer function describing the system
illustrated in the circuit below:
𝑅𝑅 𝐿𝐿
1
=
𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶 𝐼𝐼(𝑠𝑠) 𝐶𝐶𝐶𝐶
𝐼𝐼(𝑠𝑠)𝑅𝑅 × + 𝐿𝐿𝐿𝐿𝐿𝐿(𝑠𝑠) × + ×
𝐼𝐼(𝑠𝑠) 𝐼𝐼(𝑠𝑠) 𝐶𝐶𝐶𝐶 𝐼𝐼(𝑠𝑠)
15
1
=
𝐶𝐶𝐶𝐶 𝐶𝐶𝐶𝐶 𝐼𝐼(𝑠𝑠) 𝐶𝐶𝐶𝐶
𝐼𝐼(𝑠𝑠)𝑅𝑅 × + 𝐿𝐿𝐿𝐿𝐿𝐿(𝑠𝑠) × + ×
𝐼𝐼(𝑠𝑠) 𝐼𝐼(𝑠𝑠) 𝐶𝐶𝐶𝐶 𝐼𝐼(𝑠𝑠)
1
=
𝑅𝑅𝑅𝑅𝑅𝑅 + 𝐿𝐿𝐿𝐿𝑠𝑠 2 + 1
16
The importance of this cause-and-effect relationship is evidenced by the facility to represent the
relationship of system variables by diagrammatic means. The block diagram representation of the
system relationships is very common in control system engineering.
Consider an example of liquid level control system where the supply of incoming liquid is regulated by
a pneumatic valve controlled by a controller. The water level sensor provides the feedback to the
system and the outgoing liquid valve is operated manually (as needed). The schematic of the system
described is as below
Pneumatic Controller
Valve
Water Level
Sensor
Tank
Controller Valve
Sensor
As shown above, the control system block diagram consists of following elements
17
• Arrows: Represents the flow of process/logic
𝐵𝐵(𝑠𝑠)
𝐻𝐻(𝑠𝑠)
To evaluate transfer function of the closed loop shown in Fig. 2-3, a ratio of 𝐶𝐶(𝑠𝑠) with 𝑅𝑅(𝑠𝑠) is required
in terms of transfer function of forward and feedback loop.
where negative sign is used for negative feedback loop and positive sign is used for positive feedback
loop.
Similarly, using the definition of transfer function, expressions for 𝐶𝐶(𝑠𝑠) and 𝐵𝐵(𝑠𝑠) can be evaluated
Therefore,
Since,
Then,
𝐶𝐶(𝑠𝑠)
𝐸𝐸(𝑠𝑠) = (2.5)
𝐺𝐺(𝑠𝑠)
18
𝐶𝐶(𝑠𝑠)
= 𝑅𝑅(𝑠𝑠) ∓ 𝐶𝐶(𝑠𝑠)𝐻𝐻(𝑠𝑠)
𝐺𝐺(𝑠𝑠)
𝐶𝐶(𝑠𝑠)𝐻𝐻(𝑠𝑠)
= 𝑅𝑅(𝑠𝑠) �1 ∓ �
𝑅𝑅(𝑠𝑠)
𝐶𝐶(𝑠𝑠) 𝐶𝐶(𝑠𝑠)𝐻𝐻(𝑠𝑠)
= 𝐺𝐺(𝑠𝑠) �1 ∓ �
𝑅𝑅(𝑠𝑠) 𝑅𝑅(𝑠𝑠)
𝐶𝐶(𝑠𝑠) 𝐶𝐶(𝑠𝑠)𝐻𝐻(𝑠𝑠)𝐺𝐺(𝑠𝑠)
± = 𝐺𝐺(𝑠𝑠)
𝑅𝑅(𝑠𝑠) 𝑅𝑅(𝑠𝑠)
𝐶𝐶(𝑠𝑠)
[1 ± 𝐻𝐻(𝑠𝑠)𝐺𝐺(𝑠𝑠)] = 𝐺𝐺(𝑠𝑠)
𝑅𝑅(𝑠𝑠)
𝐶𝐶(𝑠𝑠) 𝐺𝐺(𝑠𝑠)
= 𝐺𝐺𝑒𝑒 (𝑠𝑠) = (2.6)
𝑅𝑅(𝑠𝑠) 1 ± 𝐺𝐺(𝑠𝑠)𝐻𝐻(𝑠𝑠)
where, in denominator, positive sign is used for negative feedback loop and negative sign is used for
positive feedback loop.
Combining
𝑋𝑋1 𝑋𝑋2 𝑋𝑋3 𝑋𝑋1 𝑋𝑋3
blocks in 𝐺𝐺1 𝐺𝐺2 𝐺𝐺1 ⋅ 𝐺𝐺2
cascade
𝑋𝑋1 𝑋𝑋2
Moving a 𝑋𝑋1 𝑋𝑋2 𝐺𝐺
pickoff point 𝐺𝐺
ahead of the 𝑋𝑋1 1
𝑋𝑋1
block 𝐺𝐺
19
𝑋𝑋1 + 𝑋𝑋3
Moving a 𝑋𝑋1 + 𝑋𝑋3 𝐺𝐺
summing point 𝐺𝐺 ±
±
1 𝑋𝑋2
ahead of a 𝑋𝑋2
block 𝐺𝐺
𝑋𝑋1 + 𝑋𝑋2
𝐺𝐺
Eliminating a 𝑋𝑋1 𝐺𝐺 𝑋𝑋2
±
feedback loop 1 ± 𝐺𝐺𝐺𝐺
𝐻𝐻
𝐻𝐻2
+ + − +
𝑅𝑅(𝑠𝑠) 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝑌𝑌(𝑠𝑠)
− +
𝐻𝐻1
𝐻𝐻3
1. The take-off point between 𝐺𝐺3 and 𝐺𝐺4 is moved ahead of 𝐺𝐺4 and 𝐻𝐻2 is
replaced by 𝐻𝐻2 /𝐺𝐺4 .
20
𝐻𝐻2
𝐺𝐺4
+ + − +
𝑅𝑅(𝑠𝑠) 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝑌𝑌(𝑠𝑠)
− +
𝐻𝐻1
𝐻𝐻3
2. The blocks 𝐺𝐺3 , 𝐺𝐺4 and 𝐻𝐻1 form a closed feedback loop and therefore
transformed using ‘feedback loop elimination’ rule.
𝐻𝐻2
𝐺𝐺4
+ + − 𝐺𝐺3 𝐺𝐺4
𝑅𝑅(𝑠𝑠) 𝐺𝐺1 𝐺𝐺2 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 𝑌𝑌(𝑠𝑠)
−
𝐻𝐻3
3. The block 𝐺𝐺2 can be combined with the block on its right using
cascading rule, and whole loop can be eliminated as in previous step.
Therefore, the equivalent transform function is given by:
𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐻𝐻2
𝐺𝐺𝑒𝑒 = ÷ �1 + × �
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 𝐺𝐺4
𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
= ÷� �
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1
𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1
= ×
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
𝐺𝐺2 𝐺𝐺3 𝐺𝐺4
=
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
Hence,
𝐻𝐻3
21
4. Finally, the last feedback loop can also be reduced using process
similar to the employed used in step 3
𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐻𝐻3
𝐺𝐺𝑒𝑒 = ÷ �1 + �
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 + 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐻𝐻3
= ÷� �
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2
= � �
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 + 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐻𝐻3
𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4
=
1 − 𝐺𝐺3 𝐺𝐺4 𝐻𝐻1 + 𝐺𝐺2 𝐺𝐺3 𝐻𝐻2 + 𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 𝐻𝐻3
These elements are shown in Figures 2-4(a) and (b), respectively. A system is represented by a line
with an arrow showing the direction of signal flow through the system. Adjacent to the line we write
the transfer function. A signal is a node with the signal’s name written adjacent to the node. Figure 2-
4(c) shows the interconnection of the systems and the signals. Each signal is the sum of signals flowing
into it. For example, 𝑉𝑉(𝑠𝑠) = 𝑅𝑅1 (𝑠𝑠)𝐺𝐺1 (𝑠𝑠) − 𝑅𝑅2 (𝑠𝑠)𝐺𝐺2 (𝑠𝑠) + 𝑅𝑅3 (𝑠𝑠)𝐺𝐺3 (𝑠𝑠) . Similarly, 𝐶𝐶2 (𝑠𝑠) =
𝑉𝑉(𝑠𝑠)𝐺𝐺5 (𝑠𝑠) = 𝑅𝑅1 (𝑠𝑠)𝐺𝐺1 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠) − 𝑅𝑅2 (𝑠𝑠)𝐺𝐺2 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠) + 𝑅𝑅3 (𝑠𝑠)𝐺𝐺3 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠).
R1(s) C1(s)
G1(s) G4(s)
–G2(s) G5(s)
R2(s) C2(s)
G3(s) V(s) –G6(s)
G(s)
R3(s) C3(s)
V(s)
(a) (b) (c)
Figure 2-4 Signal flow graph components: (a) system; (b) signal; (c)
interconnection of systems and signals
22
• Path is a branch or a continuous sequence of branches that can be traversed from one signal
(node) to another signal (node)
• Loop is a closed path that originates and terminates on the same node, with no node being
met twice along the path
• Two loops are said to be non-touching (loops) if they do not have a common node. Conversely
two loops with a common node are known as touching loops.
G6(s)
H1(s) H2(s)
H3(s)
Figure 2-5 Signal flow graph
• Loop gain. The product of branch gains found by traversing a path of a loop, referring to figure
2-5, following gains can be evaluated
𝐺𝐺2 (𝑠𝑠)𝐻𝐻1 (𝑆𝑆)
𝐺𝐺4 (𝑠𝑠)𝐻𝐻2 (𝑠𝑠)
𝐺𝐺4 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠)𝐻𝐻3 (𝑠𝑠)
𝐺𝐺4 (𝑠𝑠)𝐺𝐺6 (𝑠𝑠)𝐻𝐻3 (𝑠𝑠)
• Forward-path gain. The product of gains found by traversing a path from the input node to
the output node of the signal-flow graph in the direction of signal flow. From figure 2-5,
following forward path gains can be evaluated
𝐺𝐺1 (𝑠𝑠)𝐺𝐺2 (𝑠𝑠)𝐺𝐺3 (𝑠𝑠)𝐺𝐺4 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠)𝐺𝐺7 (𝑠𝑠)
𝐺𝐺1 (𝑠𝑠)𝐺𝐺2 (𝑠𝑠)𝐺𝐺3 (𝑠𝑠)𝐺𝐺4 (𝑠𝑠)𝐺𝐺6 (𝑠𝑠)𝐺𝐺7 (𝑠𝑠)
• Non-touching loop gain. The product of loop gains from non-touching loops taken two, three,
four, or more at a time.
• Referring to figure 2-5, non-touching loops taken two at a time are:
[𝐺𝐺2 (𝑠𝑠)𝐻𝐻1 (𝑠𝑠)][𝐺𝐺4 (𝑠𝑠)𝐻𝐻2 (𝑠𝑠)]
[𝐺𝐺2 (𝑠𝑠)𝐻𝐻1 (𝑠𝑠)][𝐺𝐺4 (𝑠𝑠)𝐺𝐺5 (𝑠𝑠)𝐻𝐻3 (𝑠𝑠)]
[𝐺𝐺2 (𝑠𝑠)𝐻𝐻1 (𝑠𝑠)][𝐺𝐺4 (𝑠𝑠)𝐺𝐺6 (𝑠𝑠)𝐻𝐻3 (𝑠𝑠)]
• There are no non-touching loops taken three (or more) at a time
where
23
Δ = 1 − Σloop gains + Σnontouching loop gains taken two at a time −
Σnontouching loop gains taken three at a time +
Σnontouching loop gains taken four at a time − ⋯
Δ is also known as the determinant
Δ𝑘𝑘 = Δ − Σ loop gain terms in Δ that touch the 𝑘𝑘 th forward path. In other words, Δ𝑘𝑘 is
formed by eliminating from Δ those loop gains that touch the 𝑘𝑘𝑡𝑡ℎ forward path.
Δ𝑘𝑘 is also known as cofactor of 𝑘𝑘 th path.
Take special notice of the alternating signs for the components of Δ. The following example will
illustrate the process and workflow of applying Mason’s rule.
𝐿𝐿1 𝐿𝐿2
𝑅𝑅(𝑠𝑠) 𝑌𝑌(𝑠𝑠)
𝐿𝐿3 𝐿𝐿4
𝐻𝐻6 𝐻𝐻7
SOLUTION: To make the signal graph flow analysis easier, the process is broken
down in simple steps, i.e. finding each component Mason’s rule as a separate
step.
24
Δ = 1 − (𝐿𝐿1 + 𝐿𝐿2 + 𝐿𝐿3 + 𝐿𝐿4 ) + (𝐿𝐿1 𝐿𝐿3 + 𝐿𝐿1 𝐿𝐿4 + 𝐿𝐿2 𝐿𝐿3 + 𝐿𝐿2 𝐿𝐿4 )
5. Evaluate the cofactor of the determinant
Cofactor of path 1 is evaluated by setting 𝐿𝐿1 = 𝐿𝐿2 = 0
Δ1 = 1 − (𝐿𝐿3 + 𝐿𝐿4 )
similarly, cofactor of path 2 can be evaluated by setting 𝐿𝐿3 = 𝐿𝐿4 = 0
Δ2 = 1 − (𝐿𝐿1 + 𝐿𝐿2 )
6. Apply Mason’s rule
Σ𝑘𝑘 𝑃𝑃𝑘𝑘 Δ𝑘𝑘
𝑇𝑇(𝑠𝑠) =
Δ
𝑃𝑃1 Δ1 + 𝑃𝑃2 Δ2
=
Δ
𝐺𝐺1 𝐺𝐺2 𝐺𝐺3 𝐺𝐺4 (1 − 𝐿𝐿3 − 𝐿𝐿4 ) + 𝐺𝐺5 𝐺𝐺6 𝐺𝐺7 𝐺𝐺8 (1 − 𝐿𝐿1 − 𝐿𝐿2 )
=
1 − 𝐿𝐿1 − 𝐿𝐿2 − 𝐿𝐿3 − 𝐿𝐿4 + 𝐿𝐿1 𝐿𝐿3 + 𝐿𝐿1 𝐿𝐿4 + 𝐿𝐿2 𝐿𝐿3 + 𝐿𝐿2 𝐿𝐿4
25
3 EQUIVALENT SYSTEM MODELLING
In previous chapters, electrical circuits were analysed and a relationship between input and output
was derived. The next step is to apply similar analysis on different engineering systems using relevant
principles and laws to derive a relationship between input and outputs.
The differential equation that governs the system shown in figure 3-1 is derived using Newton’s law
of motion and is given by
Now the above equation is converted to frequency domain using Laplace transform:
26
3.2 ROTATIONAL MECHANICAL SYSTEM
Rotational mechanical systems are handled the same way as translational mechanical systems, except
that torque replaces force and angular displacement replaces translational displacement. The
mechanical components for rotational systems are the same as those for translational systems, except
that the components undergo rotation instead of translation. Also notice that the term associated
with the mass is replaced by inertia.
Evaluation of equivalent system is quite straightforward, and the workflow described for torsional
systems may also be used for translational systems.
The analysis is initiated by writing the governing differential equations for different torques in the
system. The next step is to write equations defining relationship of torques with each other, followed
by substituting the relations to get the equation of applied torque and its affect to the whole system,
represented in a form similar to one of equation 3.3.
𝐵𝐵1 𝑁𝑁1
𝐽𝐽1
𝑇𝑇𝐿𝐿
𝑇𝑇2 , 𝜃𝜃2 𝐵𝐵2
𝑇𝑇 𝑇𝑇1 , 𝜃𝜃1
𝐽𝐽2 𝐿𝐿
𝑁𝑁2
27
SOLUTION: For ease of understanding, the analysis workflow is broken down
in several steps.
28
3.3.1 Force Voltage Analogy (Loop Analysis)
Consider an RLC circuit (as shown in figure 3-2), applying Kirchhoff’s loop law gives
𝑑𝑑 1
𝑣𝑣(𝑡𝑡) = 𝑖𝑖(𝑡𝑡)𝑅𝑅 + 𝑖𝑖(𝑡𝑡) × 𝐿𝐿 + � 𝑖𝑖(𝑡𝑡) 𝑑𝑑𝑑𝑑
𝑑𝑑𝑑𝑑 𝐶𝐶
𝑅𝑅 𝐿𝐿
𝑣𝑣(𝑡𝑡) 𝑖𝑖(𝑡𝑡) 𝐶𝐶
𝐼𝐼(𝑠𝑠)
𝑉𝑉(𝑠𝑠) = 𝐼𝐼(𝑠𝑠)𝑅𝑅 + 𝐿𝐿𝐿𝐿𝐿𝐿(𝑠𝑠) +
𝐶𝐶𝐶𝐶
Since it is known that
𝑑𝑑𝑑𝑑
𝑖𝑖(𝑡𝑡) =
𝑑𝑑𝑑𝑑
then the Laplace transform of 𝑖𝑖(𝑡𝑡) will give
𝐼𝐼(𝑠𝑠) = 𝑠𝑠𝑠𝑠(𝑠𝑠)
therefore,
𝑄𝑄(𝑠𝑠)
𝑉𝑉(𝑠𝑠) = 𝑠𝑠𝑠𝑠𝑠𝑠(𝑠𝑠) + 𝐿𝐿𝑠𝑠 2 𝑄𝑄(𝑠𝑠) +
𝐶𝐶
1
𝑉𝑉(𝑠𝑠) = 𝐿𝐿𝑠𝑠 2 𝑄𝑄(𝑠𝑠) + 𝑅𝑅𝑅𝑅𝑅𝑅(𝑠𝑠) + 𝑄𝑄(𝑠𝑠) (3.5)
𝐶𝐶
The Eqn 3.5 has the form similar to that of Eqn 3.2, and therefore the analogous quantities are as
follows:
𝑉𝑉(𝑠𝑠) ≡ 𝐹𝐹(𝑠𝑠)
𝐿𝐿 ≡ 𝑀𝑀
𝑅𝑅 ≡ 𝐵𝐵
1
≡ 𝑘𝑘
𝐶𝐶
29
𝑖𝑖(𝑡𝑡) 𝐿𝐿 𝑅𝑅 𝐶𝐶
𝑑𝑑𝑑𝑑(𝑡𝑡)
𝑣𝑣(𝑡𝑡) = 𝐿𝐿
𝑑𝑑𝑑𝑑
𝑣𝑣(𝑡𝑡)𝑑𝑑𝑑𝑑 = 𝐿𝐿 𝑑𝑑𝑑𝑑(𝑡𝑡)
1
𝑖𝑖(𝑡𝑡) = � 𝑣𝑣(𝑡𝑡) 𝑑𝑑𝑑𝑑
𝐿𝐿
Hence
1
𝐼𝐼𝐿𝐿 = � 𝑣𝑣(𝑡𝑡) 𝑑𝑑𝑑𝑑
𝐿𝐿
and
1 𝑣𝑣(𝑡𝑡) 𝑑𝑑𝑑𝑑(𝑡𝑡)
𝐼𝐼(𝑡𝑡) = � 𝑣𝑣(𝑡𝑡) 𝑑𝑑𝑑𝑑 + + 𝐶𝐶 ×
𝐿𝐿 𝑅𝑅 𝑑𝑑𝑑𝑑
Applying Laplace transform gives
1 𝑉𝑉(𝑠𝑠) 1
𝐼𝐼(𝑠𝑠) = + 𝑉𝑉(𝑠𝑠) + 𝐶𝐶𝐶𝐶𝐶𝐶(𝑠𝑠)
𝐿𝐿 𝑠𝑠 𝑅𝑅
The voltage of a circuit can be represented as a function of time as follows:
𝑑𝑑𝑑𝑑
𝑣𝑣(𝑡𝑡) =
𝑑𝑑𝑑𝑑
where 𝜙𝜙 = Magnetic Flux. Applying Laplace transform will result in following expression
𝑉𝑉(𝑠𝑠) = 𝑠𝑠𝑠𝑠(𝑠𝑠)
Hence
1 1
𝐼𝐼(𝑠𝑠) = 𝐶𝐶𝑠𝑠 2 𝜙𝜙(𝑠𝑠) + 𝑠𝑠𝑠𝑠(𝑠𝑠) 𝜙𝜙(𝑠𝑠) (3.6)
𝑅𝑅 𝐿𝐿
The Eqn 3.6 has the form similar to that of Eqn 3.2, and therefore the analogous quantities are as
follows:
𝐼𝐼(𝑠𝑠) ≡ 𝐹𝐹(𝑠𝑠)
𝐶𝐶 ≡ 𝑀𝑀
30
1
≡ 𝐵𝐵
𝑅𝑅
1
≡ 𝑘𝑘
𝐿𝐿
Mechanical Electrical
Translational Torsional Force – Voltage Force – Current
31
EXAMPLE 3.2 EQUIVALENT MECHANICAL AND ANALOGOUS ELECTRICAL SYSTEMS
PROBLEM: For the given mechanical system in the figure below, draw equivalent
mechanical system. Hence, obtain electrical analogous circuit for the given system.
𝑘𝑘
𝐵𝐵1
𝑀𝑀2
𝑥𝑥2
𝐵𝐵2
𝑀𝑀1
𝑥𝑥1
𝐹𝐹(𝑡𝑡)
SOLUTION:
Mechanical Equivalent System
To draw mechanical equivalent system, the whole system is analysed with respect
to its displacement vectors. Each component directly affected by or connected to
the displacement vector is drawn in parallel arrangement (directly connected) to its
respective displacement
In the system above, there are two displacement vectors.
• Vector 𝑥𝑥1 is directly affects (or gets affected by)
the disturbing force 𝐹𝐹(𝑡𝑡),
Mass 𝑀𝑀1 , and
Dampers 𝐵𝐵1 and 𝐵𝐵2
• Vector 𝑥𝑥2 is directly affects (or gets affected by)
Spring 𝑘𝑘,
Damper 𝐵𝐵2 and
Mass 𝑀𝑀2
Note that in the system described above, the damper 𝐵𝐵2 is common to both 𝑥𝑥1 and
𝑥𝑥2 , this should be reflected in the equivalent diagram.
32
𝑥𝑥1 𝑥𝑥2
𝐵𝐵2
𝑘𝑘
𝐹𝐹 𝑀𝑀1 𝐵𝐵1 𝑀𝑀2
The above system is described by two equations since there are two displacements.
Also note that damper 𝐵𝐵2 is under net displacement from 𝑥𝑥1 and 𝑥𝑥2 .
𝐹𝐹(𝑠𝑠) = 𝑀𝑀1 𝑠𝑠 2 𝑋𝑋1 (𝑠𝑠) + 𝐵𝐵1 𝑠𝑠𝑋𝑋1 (𝑠𝑠) + 𝐵𝐵2 𝑠𝑠[𝑋𝑋1 (𝑠𝑠) − 𝑋𝑋2 (𝑠𝑠)]
0 = 𝑀𝑀2 𝑠𝑠 2 𝑋𝑋2 (𝑠𝑠) + 𝐵𝐵2 𝑠𝑠[𝑋𝑋2 (𝑠𝑠) − 𝑋𝑋1 (𝑠𝑠)] + 𝑘𝑘𝑋𝑋2
𝐼𝐼1 𝐼𝐼2
V 𝑅𝑅2 𝐿𝐿2
𝐶𝐶
33
Electrical Equivalent Circuit: Force – Current Analogy
Using ‘Force – Current’ analogy, an equation describing analogous system can be
written as
1 1
𝐼𝐼(𝑠𝑠) = 𝐶𝐶1 𝑠𝑠 2 𝜙𝜙1 (𝑠𝑠) + 𝑠𝑠𝜙𝜙1 (𝑠𝑠) + 𝑠𝑠[𝜙𝜙1 (𝑠𝑠) − 𝜙𝜙2 (𝑠𝑠)]
𝑅𝑅1 𝑅𝑅2
1 1
0 = 𝐶𝐶2 𝑠𝑠 2 𝜙𝜙2 (𝑠𝑠) + 𝑠𝑠[𝜙𝜙2 (𝑠𝑠) − 𝜙𝜙1 (𝑠𝑠)] + 𝜙𝜙2 (𝑠𝑠)
𝑅𝑅2 𝐿𝐿
Since 𝑠𝑠 ⋅ 𝜙𝜙(𝑠𝑠) = 𝑉𝑉(𝑠𝑠),
1 1
𝐼𝐼(𝑠𝑠) = 𝐶𝐶1 𝑠𝑠𝑉𝑉1 (𝑠𝑠) + 𝑉𝑉1 (𝑠𝑠) + [𝑉𝑉1 (𝑠𝑠) − 𝑉𝑉2 (𝑠𝑠)]
𝑅𝑅1 𝑅𝑅2
1 1
0 = 𝐶𝐶2 𝑠𝑠𝑉𝑉2 (𝑠𝑠) + [𝑉𝑉2 (𝑠𝑠) − 𝑉𝑉1 (𝑠𝑠)] + 𝐼𝐼2 (𝑠𝑠)
𝑅𝑅2 𝐿𝐿𝐿𝐿
The equivalent system can also be represented by following circuit:
𝑉𝑉1 𝑉𝑉2
𝑅𝑅2
34
𝜃𝜃𝑜𝑜 𝜃𝜃𝑖𝑖
Outlet Inlet
Water Water
𝜃𝜃
𝜃𝜃𝑜𝑜
𝜃𝜃𝑖𝑖
35
SOLUTION: Rate of heat flow with respect to time is given by
𝑑𝑑𝑑𝑑 𝜃𝜃𝑜𝑜 − 𝜃𝜃𝑖𝑖
= (E3.4.1)
𝑑𝑑𝑑𝑑 𝑅𝑅
where 𝑅𝑅 is the thermal resistance of the wall of thermometer.
The temperature rise in thermometer is given by
𝑑𝑑𝑑𝑑 𝑑𝑑𝜃𝜃𝑜𝑜
= 𝐶𝐶
𝑑𝑑𝑑𝑑 𝑑𝑑𝑑𝑑
𝑑𝑑𝜃𝜃𝑜𝑜 1 𝑑𝑑𝑑𝑑
= × (E3.4.2)
𝑑𝑑𝑑𝑑 𝐶𝐶 𝑑𝑑𝑑𝑑
Substituting 𝐸𝐸3.4.1 in 𝐸𝐸3.4.2
𝑑𝑑𝜃𝜃𝑜𝑜 1
= (𝜃𝜃 − 𝜃𝜃𝑖𝑖 )
𝑑𝑑𝑑𝑑 𝐶𝐶𝐶𝐶 𝑜𝑜
𝑑𝑑𝜃𝜃𝑜𝑜 1 𝜃𝜃𝑜𝑜 − 𝜃𝜃𝑖𝑖
= � � (E3.4.3)
𝑑𝑑𝑑𝑑 𝐶𝐶 𝑅𝑅
𝐸𝐸3.4.3 implies
1
𝑠𝑠𝜃𝜃𝑜𝑜 (𝑠𝑠) = [𝜃𝜃 (𝑠𝑠) − 𝜃𝜃𝑖𝑖 (𝑠𝑠)] (E3.4.4)
𝑅𝑅𝑅𝑅 𝑜𝑜
𝑅𝑅𝑅𝑅𝑅𝑅𝜃𝜃𝑜𝑜 (𝑠𝑠) = 𝜃𝜃𝑜𝑜 (𝑠𝑠) − 𝜃𝜃𝑖𝑖 (𝑠𝑠)
Evaluating the transfer function
𝜃𝜃𝑜𝑜 (𝑠𝑠) 1
=
𝜃𝜃𝑖𝑖 (𝑠𝑠) 1 − 𝑅𝑅𝑅𝑅𝑅𝑅
Taking 𝑅𝑅𝑅𝑅 = 𝜏𝜏,
𝜃𝜃𝑜𝑜 (𝑠𝑠) 1
=
𝜃𝜃𝑖𝑖 (𝑠𝑠) 1 − 𝜏𝜏𝜏𝜏
𝑅𝑅𝑓𝑓
𝑖𝑖𝑓𝑓
𝑅𝑅𝑎𝑎 𝐿𝐿𝑎𝑎
𝐿𝐿𝑓𝑓
Load
Armature
𝑖𝑖𝑎𝑎
𝜔𝜔, 𝜃𝜃
Inertia = 𝐽𝐽
Friction = 𝑏𝑏
SOLUTION: The DC motor is a power actuator device that delivers energy to a load,
as shown in Figure above. The DC motor converts direct current (DC) electrical
energy into rotational mechanical energy. A major fraction of the torque generated
in the rotor (armature) of the motor is available to drive an external load
36
The transfer function of the DC motor will be developed for a linear approximation
to an actual motor, and second-order effects, such as hysteresis and the voltage
drop across the brushes, will be neglected. The input voltage may be applied to the
field or armature terminals. The air-gap flux 𝜙𝜙 of the motor is proportional to the
field current, provided the field is unsaturated, so that
𝜙𝜙 = 𝐾𝐾𝑓𝑓 𝑖𝑖𝑓𝑓 (E3.5 − 1)
The torque developed by the motor is assumed to be related linearly to 𝜙𝜙 and the
armature current as follows:
𝑇𝑇𝑚𝑚 = 𝐾𝐾1 𝜙𝜙𝑖𝑖𝑎𝑎 (𝑡𝑡) = 𝐾𝐾1 𝐾𝐾𝑓𝑓 𝑖𝑖𝑓𝑓 (𝑡𝑡)𝑖𝑖𝑎𝑎 (𝑡𝑡) (E3.5 − 2)
37
Therefore, the transfer function of the motor-load combination, with 𝑇𝑇𝑑𝑑 (𝑠𝑠) = 0, is
𝜃𝜃(𝑠𝑠) 𝐾𝐾𝑚𝑚
= (E3.5 − 10)
𝑉𝑉𝑓𝑓 (𝑠𝑠) 𝑠𝑠�𝑅𝑅𝑓𝑓 + 𝐿𝐿𝑓𝑓 𝑠𝑠�(𝐽𝐽𝐽𝐽 + 𝑏𝑏)
𝜃𝜃(𝑠𝑠) 𝐾𝐾𝑚𝑚
=
𝑉𝑉𝑓𝑓 (𝑠𝑠) 𝑅𝑅𝑓𝑓 𝑏𝑏
𝑠𝑠 × � + 𝑠𝑠� ⋅ 𝐿𝐿𝑓𝑓 × �𝑠𝑠 + � ⋅ 𝐽𝐽
𝐿𝐿𝑓𝑓 𝐽𝐽
Since 𝐿𝐿𝑓𝑓 /𝑅𝑅𝑓𝑓 = 𝜏𝜏𝑓𝑓 and 𝐽𝐽/𝑏𝑏 = 𝜏𝜏𝐿𝐿
𝐾𝐾𝑚𝑚
𝜃𝜃(𝑠𝑠) �𝐿𝐿𝑓𝑓 𝐽𝐽�
=
𝑉𝑉𝑓𝑓 (𝑠𝑠) 𝑠𝑠 × � + 𝑠𝑠� × �𝑠𝑠 + 1 �
1
𝜏𝜏𝑓𝑓 𝜏𝜏𝐿𝐿
𝐾𝐾𝑚𝑚 𝐿𝐿𝑓𝑓 𝐽𝐽
� �� �
𝜃𝜃(𝑠𝑠) 𝐿𝐿𝑓𝑓 𝐽𝐽 𝑅𝑅𝑓𝑓 𝑏𝑏
=
𝑉𝑉𝑓𝑓 (𝑠𝑠) 𝑠𝑠 × �1 + 𝜏𝜏𝑓𝑓 𝑠𝑠� × (𝜏𝜏𝐿𝐿 𝑠𝑠 + 1)
𝜃𝜃(𝑠𝑠) 𝐾𝐾𝑚𝑚 ⁄𝑅𝑅𝑓𝑓 𝑏𝑏
= (E3.5 − 11)
𝑉𝑉𝑓𝑓 (𝑠𝑠) 𝑠𝑠 × �1 + 𝜏𝜏𝑓𝑓 𝑠𝑠� × (𝜏𝜏𝐿𝐿 𝑠𝑠 + 1)
38
𝑑𝑑𝑑𝑑(𝑡𝑡)
= 𝜔𝜔(𝑡𝑡)
𝑑𝑑𝑑𝑑
Therefore,
𝜔𝜔(𝑠𝑠) = 𝑠𝑠𝑠𝑠(𝑠𝑠)
Hence,
𝐾𝐾𝑚𝑚
𝜃𝜃(𝑠𝑠)[𝐽𝐽𝑠𝑠 2 + 𝑏𝑏𝑏𝑏] = [𝑉𝑉 (𝑠𝑠) − 𝐾𝐾𝑏𝑏 𝑠𝑠𝑠𝑠(𝑠𝑠)]
𝑅𝑅𝑎𝑎 + 𝐿𝐿𝑎𝑎 𝑠𝑠 𝑎𝑎
𝐾𝐾𝑚𝑚 𝑉𝑉𝑎𝑎 (𝑠𝑠)
𝜃𝜃(𝑠𝑠)(𝐽𝐽𝑠𝑠 2 + 𝑏𝑏𝑏𝑏) + 𝐾𝐾𝑏𝑏 𝐾𝐾𝑚𝑚 𝑠𝑠𝑠𝑠(𝑠𝑠) =
𝑅𝑅𝑎𝑎 + 𝐿𝐿𝑎𝑎 𝑠𝑠
𝐾𝐾𝑚𝑚 𝑉𝑉𝑎𝑎 (𝑠𝑠)
𝑠𝑠[𝐽𝐽𝑠𝑠 2 + 𝑏𝑏𝑏𝑏 + 𝐾𝐾𝑏𝑏 𝐾𝐾𝑚𝑚 ]𝜃𝜃(𝑠𝑠) =
𝑅𝑅𝑎𝑎 + 𝐿𝐿𝑎𝑎 𝑠𝑠
Then,
𝜃𝜃(𝑠𝑠) 𝐾𝐾𝑚𝑚
= (E3.5 − 17)
𝑉𝑉𝑎𝑎 (𝑠𝑠) 𝑠𝑠[(𝑅𝑅𝑎𝑎 + 𝐿𝐿𝑎𝑎 𝑠𝑠)(𝐽𝐽𝐽𝐽 + 𝑏𝑏) + 𝐾𝐾𝑏𝑏 𝐾𝐾𝑚𝑚 ]
𝑥𝑥 = Case position
Accelerometer
Case
𝑦𝑦
𝑥𝑥
𝑀𝑀
Jet Engine Spring , 𝑘𝑘 𝑏𝑏
Gap, 𝛿𝛿
Rocket Sled, 𝑀𝑀𝑠𝑠
Guide Rail
SOLUTION: The aim is to design an accelerometer with an acceptable time for the
desired measurement characteristic, that is 𝑦𝑦(𝑡𝑡) = 𝑞𝑞𝑞𝑞(𝑡𝑡), to be attained (𝑞𝑞 is a
constant). The sum of forces acting on the mass is
39
𝑀𝑀(𝑦𝑦̈ + 𝑥𝑥̈ ) + 𝑏𝑏𝑦𝑦̇ + 𝑘𝑘𝑘𝑘 = 0
𝑀𝑀𝑦𝑦̈ + 𝑏𝑏𝑦𝑦̇ + 𝑘𝑘𝑘𝑘 = −𝑀𝑀𝑥𝑥̈
(E3.6 − 1)
Therefore,
𝑀𝑀
𝑀𝑀𝑦𝑦̈ + 𝑏𝑏𝑦𝑦̇ + 𝑘𝑘𝑘𝑘 = − 𝐹𝐹(𝑡𝑡)
𝑀𝑀𝑠𝑠
Or
𝑏𝑏 𝑘𝑘 𝐹𝐹(𝑡𝑡)
𝑦𝑦̈ + 𝑦𝑦̇ + 𝑦𝑦 = − (E3.6 − 2)
𝑀𝑀 𝑀𝑀 𝑀𝑀𝑠𝑠
The selection of spring, mass and damper is such that 𝑏𝑏/𝑀𝑀 = 3, and 𝑘𝑘/𝑀𝑀 = 2. The
acceleration 𝐹𝐹(𝑡𝑡)/𝑀𝑀𝑠𝑠 is denoted by 𝑄𝑄(𝑡𝑡) and the initial conditions are taken to be
𝑦𝑦(0) = −1 and 𝑦𝑦̇ (0) = 2. Thus the Laplace transform of Eqn E3.6 − 2 is
[𝑠𝑠 2 𝑌𝑌(𝑠𝑠) − 𝑠𝑠𝑠𝑠(0) − 𝑦𝑦̇ (0)] + 3[𝑠𝑠𝑠𝑠(𝑠𝑠) − 𝑦𝑦(0)] + 2𝑌𝑌(𝑠𝑠) = −𝑄𝑄(𝑠𝑠) (E3.6 − 3)
Substituting the initial conditions gives and taking 𝑄𝑄(𝑠𝑠) = 𝑃𝑃/𝑠𝑠 , where 𝑃𝑃 is the
magnitude of the step function
𝑃𝑃
[𝑠𝑠 2 𝑌𝑌(𝑠𝑠) + 𝑠𝑠 − 2] + 3[𝑠𝑠𝑠𝑠(𝑠𝑠) + 1] + 2𝑌𝑌(𝑠𝑠) = − (E3.6 − 4)
𝑠𝑠
(𝑠𝑠 2 + 𝑠𝑠 + 𝑃𝑃)
(𝑠𝑠 2 + 3𝑠𝑠 + 2)𝑌𝑌(𝑠𝑠) = −
𝑠𝑠
(𝑠𝑠 2 + 𝑠𝑠 + 𝑃𝑃)
𝑌𝑌(𝑠𝑠) = −
𝑠𝑠(𝑠𝑠 2 + 3𝑠𝑠 + 2)
(𝑠𝑠 2 + 𝑠𝑠 + 𝑃𝑃)
𝑌𝑌(𝑠𝑠) = − (𝐸𝐸3.6 − 5)
𝑠𝑠(𝑠𝑠 + 1)(𝑠𝑠 + 2)
𝐴𝐴 𝐵𝐵 𝐶𝐶
𝑌𝑌(𝑠𝑠) = + +
𝑠𝑠 𝑠𝑠 + 1 𝑠𝑠 + 2
40
When 𝑠𝑠 = −1 When 𝑠𝑠 = −2 When 𝑠𝑠 = 0
𝐵𝐵 = 𝑃𝑃 𝑃𝑃 + 2 𝑃𝑃
𝐶𝐶 = − 𝐴𝐴 = −
2 2
𝑃𝑃 𝑃𝑃 𝑃𝑃 + 2
𝑌𝑌(𝑠𝑠) = − + − (E3.6 − 6)
2𝑠𝑠 𝑠𝑠 + 1 2(𝑠𝑠 + 2)
1
𝑦𝑦(𝑡𝑡) = [−𝑃𝑃 − 2𝑃𝑃𝑒𝑒 −𝑡𝑡 − (𝑃𝑃 + 2)𝑒𝑒 −2𝑡𝑡 ], 𝑡𝑡 ≥ 0
2
41
4 TIME RESPONSE OF A SYSTEM
After the engineer obtains a mathematical representation of a subsystem, the subsystem is analysed
for its transient and steady-state responses to see if these characteristics yield the desired behaviour.
This chapter is devoted to the analysis of system transient response.
4.1 OVERVIEW
The response given by the system, which is a function of time, to an applied excitation is known as
time response.
The variation in control system’s output during the time it takes to achieve its final value is known as
transient response. It is usually characterised by an exponential or an oscillating value. The duration
of transient response is specified by the time constant of the sensor or the system (denoted by τ).
The part of the time response that remains after the complete transient response is known as steady
state response. In other words, the part of the output after the output signal has stabilized is called
steady state response. If 𝐶𝐶𝑇𝑇 (𝑡𝑡), 𝐶𝐶𝑠𝑠𝑠𝑠 (𝑡𝑡) and 𝐶𝐶𝑡𝑡 (𝑡𝑡) are the total response, steady state response and the
transient response, respectively, then
Steady state error is the difference between desired output and the steady state response.
A system that gives bounded output for bounded input is known as stable system, while systems giving
infinite outputs are known as unstable systems.
42
4.2 1ST ORDER SYSTEMS
First order control systems are characterized by first derivative with respect to time. In theory, first
order system is a system which has one integrator. As number of order increases, number of
integrators in a system also increases.
There are different techniques to solve a system equation using differential equations or Laplace
Transform but engineers find the ways to minimize the technique of solving equations for abrupt
output and work efficiency. The total response of the system is the sum of forced response and natural
response.
where
Consider a first order system and the transfer function of a system with feedback loop is given by
𝐶𝐶(𝑠𝑠) 𝐺𝐺(𝑠𝑠)
=
𝑅𝑅(𝑠𝑠) 1 + 𝐺𝐺(𝑠𝑠)𝐻𝐻(𝑠𝑠)
Moreover, as shown in the previous chapter, the transfer function of first order system in frequency
domain is always of the form
1
𝐺𝐺(𝑠𝑠) =
𝑠𝑠𝑠𝑠
Hence, taking 𝐻𝐻(𝑠𝑠) = 1 yields
1
𝐶𝐶(𝑠𝑠) 𝑠𝑠𝑠𝑠
=
𝑅𝑅(𝑠𝑠) 1 + 1 × 1
𝑠𝑠𝑠𝑠
1
=
𝑠𝑠𝑠𝑠 + 1
𝑠𝑠𝑠𝑠 � �
𝑠𝑠𝑠𝑠
1
=
𝑠𝑠𝑠𝑠 + 1
43
1 1
𝐶𝐶(𝑠𝑠) = ×
𝑠𝑠𝑠𝑠 + 1 𝑠𝑠
1
=
𝑠𝑠(𝑠𝑠𝑠𝑠 + 1)
1 𝐴𝐴 𝐵𝐵
≡ +
𝑠𝑠(𝑠𝑠𝑠𝑠 + 1) 𝑠𝑠 𝑠𝑠𝑠𝑠 + 1
𝐴𝐴(𝑠𝑠𝑠𝑠 + 1) + 𝐵𝐵𝐵𝐵
≡
𝑠𝑠(𝑠𝑠𝑠𝑠 + 1)
1 ≡ 𝐴𝐴(𝑠𝑠𝑠𝑠 + 1) + 𝐵𝐵𝐵𝐵
When 𝑠𝑠 = −1/𝜏𝜏
When 𝑠𝑠 = 0 1
𝐵𝐵 �− � = 1
𝐴𝐴 = 1 𝜏𝜏
𝐵𝐵 = −𝜏𝜏
Therefore,
1 𝜏𝜏
𝐶𝐶(𝑠𝑠) = −
𝑠𝑠 𝑠𝑠𝑠𝑠 + 1
Converting the above expression into time domain by applying inverse Laplace transformation, and
multiplying and dividing the second term with 1/𝜏𝜏.
1 𝜏𝜏
𝐶𝐶(𝑡𝑡) = ℒ −1 � − �
𝑠𝑠 𝑠𝑠𝑠𝑠 + 1
1
1 𝜏𝜏 ×
−1
=ℒ � − 𝜏𝜏 �
𝑠𝑠 1 (𝑠𝑠𝑠𝑠 + 1)
𝜏𝜏
1 1
= ℒ −1 � − �
𝑠𝑠 𝑠𝑠 + 1
𝜏𝜏
𝐶𝐶(𝑡𝑡) = 1 − 𝑒𝑒 −𝑡𝑡/𝜏𝜏
where 𝜏𝜏 is the time constant of the system. When the time elapsed is equal to the time response, i.e.
𝑡𝑡 = 𝜏𝜏, then
𝐶𝐶(𝑡𝑡) = 1 − 𝑒𝑒 −1
= 0.6321
= 63.21%
Therefore, the time constant can also be defined as the time required for a system to attain 63.2% of
its final value.
44
Response of 1st Order System
C(t)
63.2%
𝜏𝜏 Time (t)
Following table provides the output response of the system for different time periods
Time Percentage
of final value
𝒕𝒕 = 𝝉𝝉 63.21%
𝒕𝒕 = 𝟐𝟐𝟐𝟐 86.46%
𝒕𝒕 = 𝟑𝟑𝟑𝟑 95.02%
𝒕𝒕 = 𝟒𝟒𝟒𝟒 𝟗𝟗𝟗𝟗. 𝟏𝟏𝟏𝟏%
𝒕𝒕 = 𝟓𝟓𝟓𝟓 99.32%
𝒕𝒕 = 𝟏𝟏𝟏𝟏𝟏𝟏 99.99%
When elapsed time is equal to 4 times the time constant, the response is within two percent error of
final value. This is known as the settling time of a system.
When the value of time elapsed goes to infinity, the response attains a final value known as steady
state value. The difference between steady state value and the desired response is known as steady
state error, and is given by
𝑡𝑡
𝑒𝑒𝑠𝑠𝑠𝑠 = lim 𝑒𝑒(𝑡𝑡) = lim 𝑒𝑒 −𝜏𝜏
𝑡𝑡→∞ 𝑡𝑡→∞
= 𝑒𝑒 −∞
=0
45
4.2.2 Response to a Unit Impulse
Consider a first order system that is given a unit impulse input, that is
1 𝑡𝑡 = 0
𝛿𝛿(𝑡𝑡) = �
0 otherwise
When Laplace transformation is applied to unit impulse, the result is unity, that is
Therefore,
1
𝐶𝐶(𝑠𝑠) = ×1
𝑠𝑠𝑠𝑠 + 1
and
1 1
𝐶𝐶(𝑠𝑠) = � �
𝜏𝜏 𝑠𝑠 + 1
𝜏𝜏
1 −𝑡𝑡/𝜏𝜏
𝐶𝐶(𝑡𝑡) = ⋅ 𝑒𝑒
𝜏𝜏
𝑟𝑟(𝑡𝑡) = 𝑡𝑡
𝑟𝑟(𝑠𝑠) = 1/𝑠𝑠 2
Therefore,
1 1 1
𝐶𝐶(𝑠𝑠) = × 2= 2
𝜏𝜏𝜏𝜏 + 1 𝑠𝑠 𝑠𝑠 (𝜏𝜏𝜏𝜏 + 1)
Therefore,
𝐴𝐴 = 1 𝐴𝐴𝐴𝐴 + 𝐵𝐵 = 0 𝐵𝐵𝐵𝐵 + 𝐶𝐶 = 0
𝐵𝐵 + 𝜏𝜏 = 0 𝐶𝐶 = 𝜏𝜏 2
𝐵𝐵 = −𝜏𝜏
46
1 𝜏𝜏 𝜏𝜏 2
𝐶𝐶(𝑠𝑠) = − +
𝑠𝑠 2 𝑠𝑠 𝜏𝜏𝜏𝜏 + 1
1 𝜏𝜏 𝜏𝜏
= 2− +
𝑠𝑠 𝑠𝑠 𝑠𝑠 + 1
𝜏𝜏
Applying inverse Laplace transform to above equation
In this section, two physically meaningful specifications for second-order systems are defined. These
quantities can be used to describe the characteristics of the second-order transient response just as
time constants describe the first-order system response. The two quantities are called natural
frequency and damping ratio. Let us formally define them.
𝜁𝜁 = 0, Undamped response
𝜁𝜁 < 1, Underdamped response
𝜁𝜁 = 1, Critically damped response
𝜁𝜁 > 1, Over damped response
47
4.3.3 Response to Unit Step Input in 2nd Order System
The generalized transfer function of second order system is given by
𝐶𝐶(𝑠𝑠) 𝜔𝜔𝑛𝑛2
= 2
𝑅𝑅(𝑠𝑠) 𝑠𝑠 + 2𝜁𝜁𝜔𝜔𝑛𝑛 𝑠𝑠 + 𝜔𝜔𝑛𝑛2
𝜔𝜔𝑛𝑛2
𝐶𝐶(𝑠𝑠) = × 𝑅𝑅(𝑠𝑠)
𝑠𝑠 2 + 2𝜁𝜁𝜔𝜔𝑛𝑛 𝑠𝑠 + 𝜔𝜔𝑛𝑛2
1
𝑅𝑅(𝑠𝑠) =
𝑠𝑠
Therefore,
𝜔𝜔𝑛𝑛2 1
𝐶𝐶(𝑠𝑠) = 2 2 × 𝑠𝑠
𝑠𝑠 + 2𝜁𝜁𝜔𝜔𝑛𝑛 𝑠𝑠 + 𝜔𝜔𝑛𝑛
𝜔𝜔𝑛𝑛2 1
= 2 2 2 2 ×
𝑠𝑠 + 2𝜁𝜁𝜔𝜔𝑛𝑛 𝑠𝑠 + 𝜁𝜁 2 𝜔𝜔𝑛𝑛 − 𝜁𝜁 2 𝜔𝜔𝑛𝑛 + 𝜔𝜔𝑛𝑛 𝑠𝑠
𝜔𝜔𝑛𝑛2 1
= 2 2 2 2 2 2 × 𝑠𝑠
𝑠𝑠 + 2𝜁𝜁𝜔𝜔𝑛𝑛 𝑠𝑠 + 𝜁𝜁 𝜔𝜔𝑛𝑛 − 𝜁𝜁 𝜔𝜔𝑛𝑛 + 𝜔𝜔𝑛𝑛
𝜔𝜔𝑛𝑛2 1
= 2 ×
(𝑠𝑠 + 𝜁𝜁𝜔𝜔𝑛𝑛 ) + (1 − 𝜁𝜁 )𝜔𝜔𝑛𝑛 𝑠𝑠
2 2
𝜔𝜔𝑛𝑛2 1
= 2 ×
(𝑠𝑠 + 𝜁𝜁𝜔𝜔𝑛𝑛 ) + 𝜔𝜔𝑑𝑑 𝑠𝑠
2
Applying inverse Laplace transformation (refer to transformation 19 and 20 in the “Table of Laplace
Transforms”).
𝜁𝜁𝜔𝜔𝑛𝑛 −𝜁𝜁𝜔𝜔 𝑡𝑡
ℒ −1 [𝐶𝐶(𝑠𝑠)] = 1 − 𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡 cos(𝜔𝜔𝑑𝑑 𝑡𝑡) − 𝑒𝑒 𝑛𝑛 sin(𝜔𝜔 𝑡𝑡)
𝑑𝑑
𝜔𝜔𝑑𝑑
𝜁𝜁𝜔𝜔𝑛𝑛
= 1 − 𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡 �cos(𝜔𝜔𝑑𝑑 𝑡𝑡) − sin(𝜔𝜔𝑑𝑑 𝑡𝑡) × �
𝜔𝜔𝑛𝑛 �1 − 𝜁𝜁 2
𝜁𝜁
𝑐𝑐(𝑡𝑡) = 1 − 𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡 �cos(𝜔𝜔𝑑𝑑 𝑡𝑡) − sin(𝜔𝜔𝑑𝑑 𝑡𝑡) × �
�1 − 𝜁𝜁 2
48
Using following substitution
sin 𝜙𝜙 = �1 − 𝜁𝜁 2 cos 𝜙𝜙 = 𝜁𝜁
Then
�1 − 𝜁𝜁 2
tan 𝜙𝜙 =
𝜁𝜁
𝜁𝜁
𝑐𝑐(𝑡𝑡) = 1 − 𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡 �cos(𝜔𝜔𝑑𝑑 𝑡𝑡) − sin(𝜔𝜔𝑑𝑑 𝑡𝑡) × �
�1 − 𝜁𝜁 2
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
=1− �cos(𝜔𝜔𝑑𝑑 𝑡𝑡) �1 − 𝜁𝜁 2 − 𝜁𝜁 sin(𝜔𝜔𝑑𝑑 𝑡𝑡)�
�1 − 𝜁𝜁 2
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
=1− [cos(𝜔𝜔𝑑𝑑 𝑡𝑡) sin 𝜙𝜙 − sin(𝜔𝜔𝑑𝑑 𝑡𝑡) cos 𝜙𝜙]
�1 − 𝜁𝜁 2
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
=1− [sin(𝜔𝜔𝑑𝑑 𝑡𝑡 + 𝜙𝜙)]
�1 − 𝜁𝜁 2
The time required for output to achieve fifty percent of its final value for the first time.
Time required for response to achieve its maximum value for the first time
Derivation of 𝒕𝒕𝒑𝒑
𝑑𝑑
𝑐𝑐(𝑡𝑡) = 0
𝑑𝑑𝑑𝑑
49
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
�cos�𝜔𝜔𝑑𝑑 𝑡𝑡𝑝𝑝 + 𝜙𝜙� 𝜔𝜔𝑑𝑑 − sin�𝜔𝜔𝑑𝑑 𝑡𝑡𝑝𝑝 + 𝜙𝜙� 𝜁𝜁𝜔𝜔𝑛𝑛 � = 0
�1 − 𝜁𝜁 2
cos�𝜔𝜔𝑑𝑑 𝑡𝑡𝑝𝑝 + 𝜙𝜙� 𝜔𝜔𝑛𝑛 �1 − 𝜁𝜁 2 − sin�𝜔𝜔𝑑𝑑 𝑡𝑡𝑝𝑝 + 𝜙𝜙� 𝜁𝜁𝜔𝜔𝑛𝑛 = 0
Time required for a system to acquire its final value for the first time
Derivation of 𝒕𝒕𝒓𝒓
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
𝑐𝑐(𝑡𝑡) = 1 − [sin(𝜔𝜔𝑑𝑑 𝑡𝑡 + 𝜙𝜙)]
�1 − 𝜁𝜁 2
𝑐𝑐(𝑡𝑡) = 1
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡𝑟𝑟
[sin(𝜔𝜔𝑑𝑑 𝑡𝑡𝑟𝑟 + 𝜙𝜙)] = 0
�1 − 𝜁𝜁 2
Hence
𝜔𝜔𝑑𝑑 𝑡𝑡𝑟𝑟 + 𝜙𝜙 = 𝜋𝜋 𝑛𝑛 = 1
𝜋𝜋 − 𝜙𝜙
𝑡𝑡𝑟𝑟 =
𝜔𝜔𝑑𝑑
𝜋𝜋 − 𝜙𝜙
=
𝜔𝜔𝑛𝑛 �1 − 𝜁𝜁 2
𝑐𝑐(𝑡𝑡) − 𝑐𝑐(∞)
𝑀𝑀𝑝𝑝 = × 100%
𝑐𝑐(∞)
𝑐𝑐(𝑡𝑡) − 1
= × 100%
1
= 100 × 𝑐𝑐(𝑡𝑡)%
50
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡
=1− [sin(𝜔𝜔𝑑𝑑 𝑡𝑡 + 𝜙𝜙)]
�1 − 𝜁𝜁 2
Since maximum overshoot is the difference between final value and the first peak, this
point occurs at peak time, therefore
𝑀𝑀𝑝𝑝 = 𝑐𝑐�𝑡𝑡𝑝𝑝 �
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡𝑝𝑝𝜔𝜔𝑑𝑑 𝜋𝜋
=1− �sin � + 𝜙𝜙��
�1 − 𝜁𝜁 2 𝜔𝜔𝑑𝑑
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡𝑝𝑝
=1− [sin(𝜋𝜋 + 𝜙𝜙)]
�1 − 𝜁𝜁 2
𝑒𝑒 −𝜁𝜁𝜔𝜔𝑛𝑛 𝑡𝑡𝑝𝑝
=1+ sin(𝜙𝜙)
�1 − 𝜁𝜁 2
𝜋𝜋
−𝜁𝜁𝜔𝜔𝑛𝑛
𝑒𝑒 𝜔𝜔𝑛𝑛 �1−𝜁𝜁 2
=1+ sin(𝜙𝜙)
�1 − 𝜁𝜁 2
−𝜁𝜁𝜁𝜁
�1−𝜁𝜁 2
𝑒𝑒
=1+ �1 − 𝜁𝜁 2 sin 𝜙𝜙 = 𝜁𝜁
�1 − 𝜁𝜁 2
When the output of the system reaches within 2% or 5% of steady state value. The settling
time can be calculated from the following expressions
1
𝜏𝜏 =
𝜁𝜁𝜔𝜔𝑛𝑛
3
𝑡𝑡𝑠𝑠,5% = 3𝜏𝜏 =
𝜁𝜁𝜔𝜔𝑛𝑛
4
𝑡𝑡𝑠𝑠,2% = 4𝜏𝜏 =
𝜁𝜁𝜔𝜔𝑛𝑛
51
5 STABILITY OF SYSTEMS
Stability of closed-loop feedback systems is central to control system design. A stable system should
exhibit a bounded output if the corresponding input is bounded. This is known as bounded-input-
bounded-output stability and is one of the main topics of this chapter. The stability of a feedback
system is directly related to the location of the roots of the characteristic equation of the system
transfer function and to the location of the eigenvalues of the system matrix for a system in state
variable format. The Routh-Hurwitz method is introduced as a useful tool for assessing system
stability. The technique allows the determination of the number of roots of the characteristic equation
in the right half plane without actually computing the values of the roots. This provides a design
method for determining values of certain system parameters that will lead to closed-loop stability.
The concept of stability can be illustrated by considering a right circular cone placed on a plane
horizontal surface. If the cone is resting on its base and is tipped slightly, it returns to its original
equilibrium position. This position and response are said to be stable. If the cone rests on its side and
is displaced slightly, it rolls with no tendency to leave the position on its side. This position is
designated as the neutral stability. On the other hand, if the cone is placed on its tip and released, it
falls onto its side. This position is said to be unstable.
The stability of a dynamic system is defined in a similar manner. The response to a displacement, or
initial condition, will result in either a decreasing, neutral, or increasing response.
The location in the s-plane of the poles of a system indicates the resulting transient response. The
poles in the left-hand portion of the s-plane result in a decreasing response for disturbance inputs.
Similarly, poles on the 𝑗𝑗𝑗𝑗 − axis and in the right-hand plane result in a neutral and an increasing
response, respectively, for a disturbance input.
In terms of linear systems, the stability requirement may be defined in terms of the location of the
poles of the closed-loop transfer function. The closed-loop system’s generalised transfer function in
frequency domain is written as
To obtain a bounded response, the poles of the closed-loop system must be in the left-hand portion
of the s-plane. Thus,
52
A necessary and sufficient condition for a feedback system to be stable is that
all the poles of the system transfer function have negative real parts
A system is stable if all the poles of the transfer function are in the left-hand s-plane. A system is not
stable if not all the roots are in the left-hand plane. If the characteristic equation has simple roots on
the imaginary axis (𝑗𝑗𝑗𝑗 − axis) with all other roots in the left half-plane, the steady-state output will
be sustained oscillations for a bounded input, unless the input is a sinusoid (which is bounded) whose
frequency is equal to the magnitude of the 𝑗𝑗𝑗𝑗 − axis roots. For this case, the output becomes
unbounded. Such a system is called marginally stable, since only certain bounded inputs (sinusoids of
the frequency of the poles) will cause the output to become unbounded. For an unstable system, the
characteristic equation has at least one root in the right half of the s-plane or repeated 𝑗𝑗𝑗𝑗 roots; for
this case, the output will become unbounded for any input.
1. Generate a data table called a Routh table (also known as Routh’s Array)
2. Interpret the Routh table to tell how many closed-loop system poles are in the left half-plane,
the right half-plane, and on the 𝑗𝑗𝑗𝑗-axis.
The power of the method lies in design rather than analysis. For example, if there is an unknown
parameter in the denominator of a transfer function, it is difficult to determine via a calculator the
range of this parameter to yield stability, requires tedious trial and error method to answer the
stability question. Meanwhile, Routh-Hurwitz criterion can yield a closed-form expression for the
range of the unknown parameter.
53
𝑏𝑏1 𝑎𝑎2𝑛𝑛+1 − 𝑏𝑏𝑛𝑛+1 𝑎𝑎1
𝑐𝑐𝑛𝑛 =
𝑏𝑏1
For a system to be stable, the elements of the first column of Routh’s Array
must have the same sign (either whole column is positive, or the whole column
is negative)
The number instances when the sign change occurs in the first column is equal to number of poles in
the right half of the s – plane.
𝑠𝑠 5 1 3 2
𝑠𝑠 4 2 6 1
𝑠𝑠 3 𝜖𝜖 1.5 0
6𝜖𝜖 − 3
𝑠𝑠 2 1 0
𝜖𝜖
6𝜖𝜖 − 3
1.5 � � − 𝜖𝜖
𝜖𝜖
𝑠𝑠1
6𝜖𝜖 − 3
� �
𝜖𝜖
𝑠𝑠 0
Therefore, when 𝜖𝜖 → 0
6𝜖𝜖 − 3 6𝜖𝜖 3 3
lim � � = lim � − � = lim �6 − � = lim (6 − ∞) = −∞
𝜖𝜖→0 𝜖𝜖 𝜖𝜖→0 𝜖𝜖 𝜖𝜖 𝜖𝜖→0 𝜖𝜖 𝜖𝜖→0
Similarly,
6𝜖𝜖 − 3
1.5 � � − 𝜖𝜖 1.5(−∞) − 0
lim � 𝜖𝜖 �= = 1.5
𝜖𝜖→∞ 6𝜖𝜖 − 3 −∞
� �
𝜖𝜖
Hence, the Routh’s Array can be written as
54
𝑠𝑠 5 1 3 2
𝑠𝑠 4 2 6 1
𝑠𝑠 3 𝜖𝜖 1.5 0
𝑠𝑠 2 −∞ 1 0
𝑠𝑠1 1.5
𝑠𝑠 0
As shown above, there is a sign change in 4th row, therefore the system is unstable.
This case can be tackled by forming an auxiliary equation from the row above and differentiating to
form the equation for the row below. The following illustrative example will better demonstrate the
process.
𝑠𝑠 5 1 6 8
𝑠𝑠 4 7 42 56
𝑠𝑠 3 0 0 0
The auxiliary equation from the second row is given by: 7𝑠𝑠 4 + 42𝑠𝑠 2 + 56
𝑠𝑠 5 1 6 8
𝑠𝑠 4 7 42 56
0 0 0
𝑠𝑠 3
28 84 0
𝑠𝑠 2 21 56 0
KG(𝑠𝑠)𝐻𝐻(𝑠𝑠) = 0
2. Evaluate singularities (poles and zeros) of the transfer function and plot the points
• The path of the locus is such that it goes from pole to zero
55
• Each pole must be pared with a unique zero
In case of excess of poles, the locus from pole seeks zero at infinity (implied
zero)
In case of excess of zeros, the locus approaches zero from infinity (implied
pole)
(2𝑘𝑘 + 1)180°
Asymptote Angle = 𝑘𝑘 = 0, 1, … , (𝑝𝑝 − 𝑧𝑧 − 1)
𝑝𝑝 − 𝑧𝑧
Σ𝑝𝑝𝑖𝑖 − Σ𝑧𝑧𝑖𝑖
Asymptote & Real Axis Intercept = cg =
𝑝𝑝 − 𝑧𝑧
5. Breakaway point (𝜎𝜎𝑏𝑏 ) refers to the intersection of locus from conjugate pole to the real axis
and is given by
either
𝑝𝑝 𝑧𝑧
1
� = �(𝜎𝜎𝑏𝑏 − 𝑧𝑧𝑖𝑖 )
(𝜎𝜎𝑏𝑏 − 𝑝𝑝𝑖𝑖 )
𝑖𝑖=1 𝑖𝑖=1
or
𝐾𝐾 = 𝑓𝑓(𝑠𝑠)
and evaluating
𝑑𝑑𝑑𝑑
=0
𝑑𝑑𝑑𝑑
6. Intersection of locus with real axis occurs at ±90°
𝜃𝜃𝑑𝑑 = − � 𝜃𝜃 + 180°
Where 𝜃𝜃 for poles is taken positive and 𝜃𝜃 for zeros are taken as negative, that is
8. Locus intersection with imaginary axis can be calculated using Routh’s Array for closed loop
transfer function. Firstly, evaluate the closed loop transfer function from the given open loop
function by using
𝐾𝐾𝐾𝐾(𝑠𝑠)
Closed Loop T. F =
1 + 𝐾𝐾𝐾𝐾(𝑠𝑠)𝐻𝐻(𝑠𝑠)
56
• Construct the Routh’s Array and look for a row that can be completely zero for a given
value of 𝐾𝐾 such that 𝐾𝐾 > 0, and find the corresponding value of 𝐾𝐾.
• After evaluating value of 𝐾𝐾 , calculate the value of 𝑠𝑠 by using the any row of the
Routh’s Array.
Im(𝑠𝑠0 )
tan 𝜃𝜃 = � �
Re(𝑠𝑠0 )
• Now using the determined Im(𝑠𝑠0 ) and the estimate of Re(𝑠𝑠0 ) , evaluate the
radius/magnitude of the complex number ‘𝑠𝑠’
2. For the correct value of ‘𝑠𝑠’, sum of negative pole angles and positive zero angle will result in
an angle that is an odd multiple of 180°, that is
𝑝𝑝 𝑧𝑧
• Keep varying the value of radius while keeping the angle constant until the above
equation is satisfied.
3. To find the value of gain constant 𝐾𝐾 at the value of ‘𝑠𝑠’ found in previous step, following
equation can be used
|𝐾𝐾𝐾𝐾(𝑠𝑠)𝐻𝐻(𝑠𝑠)| = 1
57