Advanced Control Systems

Lecture 3 - 4
Chapter 2: Mathematical Models of
Systems
Mohebullah Wali, Ph.D.
Faculty of Engineering, Kabul University
Department of Electrical and Electronics Engineering
Chapter 2: Mathematical Models of Systems

2.1 Introduction
2.2 Differential Equations of Physical Systems
2.3 Linear Approximations of Physical Systems
2.4 The Laplace Transform
2.5 The Transfer Function of Linear Systems
2.6 Block Diagram Models
2.7 Signal-Flow Graph Models
2.8 Design Examples
2.9 The Simulation of Systems Using Control Design Software
2.10 Sequential Design Example: Disk Drive Read System
2.11 Summary
Chapter 2: Mathematical Models of Systems
DESIRED OUTCOMES
Upon completion of Chapter 2, students should:
 Recognize that differential equations can describe the dynamic
behavior of physical systems.
 Be able to utilize linearization approximations through the use
of Taylor series expansions.
 Understand the application of Laplace transforms and their
role in obtaining transfer functions.
 Be aware of block diagrams (and signal-flow graphs) and their
role in analyzing control systems.
 Understand the important role of modeling in the control
system design process.
Chapter 2: Mathematical Models of Systems

Main Contents of Lecture 3

 Introduction to mathematical models of systems

 Differential equations of physical systems
 Linear approximations of physical systems
Chapter 2: Mathematical Models of Systems

Main Contents of Lecture 3

 Introduction to mathematical models of systems

 Differential equations of physical systems
 Linear approximations of physical systems
1.1 Introduction to Mathematical Models of Systems

• For the analysis and design of control systems, quantitative

mathematical models of systems must be obtained
• Descriptive equations for dynamic systems are DE’s

• Linear approximations can be utilized to linearize the DE’s and

use the Laplace transform to simplify the solution

• To simplify the complexity of system, necessary assumptions

can be considered
1.1 Introduction to Mathematical Models of Systems

In summary, the approach to dynamic system modeling can be

listed as follows:

1. Define the system and its components.

2. Formulate the mathematical model and fundamental necessary
assumptions based on basic principles.
3. Obtain the differential equations representing the mathematical
model.
4. Solve the equations for the desired output variables.
5. Examine the solutions and the assumptions.
6. If necessary, reanalyze or redesign the system.
Chapter 2: Mathematical Models of Systems

Main Contents of Lecture 3

 Introduction to mathematical models of systems

 Differential equations of physical systems
 Linear approximations of physical systems
2.2 Differential Equations of Physical Systems

• The differential equations describing the dynamic

performance of a physical system are obtained by
utilizing the physical laws of the process
• The approach applies equally to mechanical,
electrical, fluid, and thermodynamic systems
• Examples on next slides:
– RLC circuit
– Torsional Spring-Mass system
2.2 Differential Equations of Physical Systems

RLC Circuit

𝑉𝐿 + 𝑉𝑅 + 𝑉𝐶 = 𝐸(𝑡)

𝑑𝑖 1
𝐿 + 𝑅𝑖 + න 𝑖𝑑𝑡 = 𝐸 𝑡
𝑑𝑡 𝐶

𝑑2 𝑖 𝑑𝑖 1
𝐿 2 + 𝑅 + 𝑖 = 𝐸(𝑡)
𝑑𝑡 𝑑𝑡 𝐶
2.2 Differential Equations of Physical Systems

Torsional Spring-Mass System

• Ta is applied toque
• Assume the torsional spring
element is massless
• It is needed to measure the torque
Ts transmitted to the mass m
• The sum of the torques acting on
the spring itself must be zero,

𝑇𝑎 (𝑡) − 𝑇𝑠 (𝑡) = 0 → 𝑇𝑎 (𝑡) = 𝑇𝑠 (𝑡)

• The external torque Ta is transmitted through the torsional spring
• Torque is a through-variable
2.2 Differential Equations of Physical Systems

Torsional Spring-Mass System

• The angular rate difference
associated with torsional spring
element is
𝜔 𝑡 = 𝜔𝑠 (𝑡) − 𝜔𝑎 (𝑡)
• The angular rate difference is
measured across the torsional
spring element and is referred to
as an across-variable

• These same types of arguments can be made for most common

physical variables such as,
force, current, volume, flow rate, etc.
2.2 Differential Equations of Physical Systems
Summary of through- and across-variables of dynamic
systems
2.2 Differential Equations of Physical Systems
2.2 Differential Equations of Physical Systems
2.2 Differential Equations of Physical Systems
2.2 Differential Equations of Physical Systems

Nomenclature
Through-variable: Capacitive storage:
F = force,
C = capacitance,
T = torque,
I = current, M = mass,
Q = fluid volumetric flow rate, J = moment of inertia,
q = heat flow rate Cf = fluid capacitance,
Across-variable: Ct = thermal capacitance
v = translational velocity,
ω = angular velocity, v = voltage, Energy dissipators:
P = pressure, R = resistance,
ℱ = temperature b = viscous friction,
Inductive storage: Rf = fluid resistance,
L = inductive, Rt = thermal resistance
1/k = reciprocal translational or
rotational stiffness,
I = fluid inertance
2.2 Differential Equations of Physical Systems

• For mechanical systems, Newton’s laws are used

• For electrical systems, Kirchhoff’s voltage laws
are used
 2.2 Differential Equations of Physical Systems
Mass-Spring-Damper System
• Three types of forces that resist the
motion:
– Inertia force of moving body
𝑑2𝑦
𝑓𝑀 = 𝑀𝑎 = 𝑀 2
𝑑𝑡

– Damping force due to viscous

friction
𝑑𝑦
𝑓𝑑 = 𝑏
𝑑𝑡

– And spring force

𝑓𝑠 = 𝑘y

• Summing the forces acting on M:

𝑑 2 𝑦(𝑡) 𝑑𝑦(𝑡)
𝑀 +𝑏 + 𝑘𝑦(𝑡) = 𝑟(𝑡)
𝑑𝑡 2 𝑑𝑡
Second-order linear constant-coefficient differential equation
 2.2 Differential Equations of Physical Systems
RLC Circuit
• Current in the braches:
– Current in resistor
𝑣(𝑡)
𝑖𝑅 =
𝑅
– Current in inductor
1 𝑡
𝑖𝐿 = ‫׬‬0 𝑣 𝑡 𝑑𝑡
𝐿
– Current in capacitor
𝑑𝑣(𝑡)
𝑖𝐶 = 𝐶
𝑑𝑡

• By Kirchoff’s current law:

𝑣(𝑡) 𝑑𝑣(𝑡) 1 𝑡
+𝐶 + න 𝑣 𝑡 𝑑𝑡 = 𝑟(𝑡)
𝑅 𝑑𝑡 𝐿 0
Integrodifferential equation
2.2 Differential Equations of Physical Systems
Similarity of mechanical and electrical systems
• The solution of DEs can be obtained by the use of integrating
factor and the method of undetermined coefficients:
• Solution of Mass-Spring-Damper DE when mass is initially
displaced a distance y(0) = y0 and released
𝑑 2 𝑦(𝑡) 𝑑𝑦(𝑡)
𝑀 +𝑏 + 𝑘𝑦(𝑡) = 𝑟(𝑡)
𝑑𝑡 2 𝑑𝑡

y t = 𝐾1 𝑒 −𝛼1𝑡 sin(𝛽1 𝑡 + 𝜃1 )
• Solution of RLC Circuit when r(t) = I (constant current)
𝑣(𝑡) 𝑑𝑣(𝑡) 1 𝑡
+𝐶 + ‫׬‬0 𝑣 𝑡 𝑑𝑡 = 𝑟(𝑡)
𝑅 𝑑𝑡 𝐿

𝑣 𝑡 = 𝐾2 𝑒 −𝛼2 𝑡 cos(𝛽2 𝑡 + 𝜃2 )
2.2 Differential Equations of Physical Systems

• Typical voltage response for underdamped RLC circuit.

 2.2 Differential Equations of Physical Systems
Further close similarity between mechanical and electrical systems
𝑑𝑦
• The mass-spring-damper DE in terms of velocity: 𝑣 𝑡 =
𝑑𝑡
2
𝑑 𝑦(𝑡) 𝑑𝑦(𝑡)
𝑀 2
+𝑏 + 𝑘𝑦(𝑡) = 𝑟(𝑡)
𝑑𝑡 𝑑𝑡

𝑡
𝑑𝑣(𝑡)
𝑀 + 𝑏𝑣(𝑡) + 𝑘 න 𝑣 𝑡 𝑑𝑡 = 𝑟(𝑡)
𝑑𝑡 0
Which is similar to RLC circuit DE

𝑑𝑣(𝑡) 1 1 𝑡
𝐶 + 𝑣(𝑡) + න 𝑣 𝑡 𝑑𝑡 = 𝑟(𝑡)
𝑑𝑡 𝑅 𝐿 0
• Velocity v(t) and voltage v(t) are equivalent variables, called
analogous variables and the systems are analogous systems.
• This voltage-velocity analogy, often called force-current analogy.
Chapter 2: Mathematical Models of Systems

Main Contents of Lecture 3

 Introduction to mathematical models of systems

 Differential equations of physical systems
 Linear approximations of physical systems
 2.3 Linear Approximations of Physical Systems
• A great majority of physical systems are linear within some range
of the variables
• Systems ultimately become nonlinear as the variables are increased
without limit.
• The spring-mass-damper system in previous slide is linear as long
as the mass is subjected to small deflections y(t). If y(t) are
continually increased, eventually the spring will be overextended
and break.
• A system is defined as linear in terms of the system excitation and
response.
– Electrical network, excitation is the input current r(t) and the
response is the voltage v(t)
 2.3 Linear Approximations of Physical Systems
• A system is linear if it satisfy the Principle of Superposition and
Homogeneity
• Principle of Superposition:

x1(t) y1(t) x2(t) y2(t)

System System

x1(t)+x2(t) y1(t)+y2(t)
System

Principle of Superposition is satisfied

2.3 Linear Approximations of Physical Systems
• Homogeneity

ax1(t) ay1(t) bx2(t) by2(t)

System System

ax1(t)+bx2(t) ay1(t)+by2(t)
System

Homogeneity is satisfied
A linear system satisfies the properties of
superposition and homogeneity.
 2.3 Linear Approximations of Physical Systems
• A system characterized by the relation y = x2 is not linear, it does
not satisfy the superposition property
• A system represented by relation y = mx + b is not linear, it does
not satisfy the homogeneity property
• The second can be considered linear about an operating point x0, y0
for small changes Δx and Δy. When x = x0 + Δx and y = y0 + Δ y,
we have
y = mx + b
or
y0 + Δ y = mx0 + mΔx + b
Therefore, Δ y = mΔx, which satisfy the necessary condition
2.3 Linear Approximations of Physical Systems
• Nonlinear elements can often be linearized assuming small-signal
conditions
• A general element represented by
y(t) = g(x(t))
The normal operating point would be designated by x0. For
linearization, a Taylor Series expansion about operating point can be
used as
′ 𝑥−𝑥0 ′′ 𝑥−𝑥0 2
𝑦 = 𝑔 𝑥 = 𝑔 𝑥0 + 𝑔 𝑥0 + 𝑔 𝑥0 + ...
1! 2!

The slope at the operating point, g’(x0) is a good approximation to

the curve over a small range of (x – x0), the derivative from the
operating point
2.3 Linear Approximations of Physical Systems
• As a reasonable approximation, the equation of Taylor series for
y = g(x) becomes

𝑦 = 𝑔 𝑥0 + 𝑔′ 𝑥0 𝑥 − 𝑥0 = 𝑦0 + 𝑚(𝑥 − 𝑥0)

where m is the slope at the operating point. We can rewrite above

equation as linear equation

𝑦 − 𝑦0 = 𝑚(𝑥 − 𝑥0)
or
Δ y = mΔx
2.3 Linear Approximations of Physical Systems
• Mass M sitting on a nonlinear spring
• The normal operating point is the
equilibrium position, when the spring
force balances the gravitational force
Mg
𝑓0 = 𝑀𝑔
• For the nonlinear spring f = y2 ,
equilibrium position is
𝑦 = (𝑓)1/2 → 𝑦0 = (𝑀𝑔)1/2
• The linear model for small deviation
is ∆𝑓 = 𝑚∆𝑦
𝑑𝑓
𝑚= ቮ
where 𝑑𝑦 𝑦
0

Thus m = 2y0
2.3 Linear Approximations of Physical Systems
• It the dependent variable y depends upon several excitation
variables x1, x2, . . . , xn, then the relationship is written as

𝑦 = 𝑔 𝑥1 , 𝑥2 , … , 𝑥𝑛
The Taylor series expansion about the operating point x10, x20, . .
xn0 is useful for linear approximation and can be written as

𝜕𝑔 𝜕𝑔
𝑦 = 𝑔 𝑥10 , 𝑥20 , … , 𝑥𝑛0 + ቮ 𝑥1 − 𝑥10 + ቮ 𝑥2 − 𝑥20
𝜕𝑥1 𝑥 = 𝑥 𝜕𝑥2 𝑥 = 𝑥
0 0

𝜕𝑔
+⋯+ ቮ 𝑥𝑛 − 𝑥𝑛0
𝜕𝑥𝑛 𝑥 = 𝑥
0
2.3 Linear Approximations of Physical Systems
Example: Pendulum oscillator model
• The torque on the mass is
𝑇 = 𝑀𝑔𝐿 sin 𝜃
• The equilibrium condition for the
mass is 𝜃0 = 0o
• The linear approximation at
equilibrium is
𝜕 sin 𝜃
𝑇 − 𝑇0 = 𝑀𝑔𝐿 ቮ 𝜃 − 𝜃0
𝜕𝜃 𝜃 = 𝜃
0

where T0 = 0. Then, we have

𝑇 = 𝑀𝑔𝐿 cos 0o 𝜃 − 0o = 𝑀𝑔𝐿𝜃

The approximation is reasonably accurate for −𝜋Τ4 ≤ 𝜃 ≤ 𝜋Τ4

End of Lecture 3 and 4