Chapter 1 – Introduction to Dynamic Systems and Control

1.1 Introduction

In solving engineering problems, there is a need to understand and determine the

dynamic response of a physical system that may consist of several components.
These efforts involve: Modelling - Analysis – Simulation. Because building a prototype and
conducting experimental tests are either infeasible or too expensive for a preliminary design.
Dynamic systems and control involve the analysis, design and control of physical
engineering systems that are often composed of interacting mechanical, electrical, and fluid
subsystems components.

1.2 Definitions

System: A combination of components acting together to perform a specified objective.

These components have input-output relationships. Example: DC motor (input: voltage – output:
angular velocity of the motor shaft).
Dynamic system: A system where the current output variables (or dynamic variables)
depend end on the initial conditions of the system and/or the previous input variables. The
dynamic variables of the system vary with time. Example: for the DC motor, the angular velocity
is the dynamic variable, and the voltage is the input.
Modeling: The process of applying the appropriate physical laws to derive the
mathematical equations that describe the physics of the system. Example: for the DC motor, the
electrical circuit is modeled by using Kirchhoff’s voltage law and the mechanical motion is
modeled by using Newton’s second law.
Mathematical model: A mathematical description of a dynamic system’s behavior which
is usually a set of differential equations. Example: for the DC motor example, the mathematical
model consists of a DE for the electrical current and a DE for the mechanical motion.
Simulation: The process of obtaining the system’s dynamic response by numerically
solving the governing modelling equations.
System analysis: The use of calculations or simulation to determine the system response
to assess its performance. Repeated analysis aids the design process where the system’s
parameters are altered to improve performance. Example: we might apply a constant voltage
input and determine the angular velocity repones calculations or simulations.
1.3 Classification of dynamic systems

Distributed vs. lumped systems

A distributed system required an infinite number of “internal” variables. A lumped system
involved a finite number of “internal” variables. Example: we would “lump” all the pressure
distribution in a cylinder champer into one single pressure term.
Continuous-time vs. discrete-time systems
A continuous-time system involved variables and function that are defined for all the time.
A discrete-time system involved variable that are defined only at discrete time inputs.
Time-varying vs. time-invariant systems
In a time-varying system, the system parameters change with time (e.g., the friction
coefficient changes with time). In a time-invariant system, the parameters remain constant.
Example: for the DC motor, the system parameters would be electrical resistant of the circuit,
inductance of the coil windings around the rotor, …..
Linear vs. nonlinear systems
Suppose a system of input-output relationship that is described by the function 𝑦 = 𝑓(𝑥)
where 𝑥 is the input and 𝑦 is the output. Linear systems obey the superposition property:
1. If 𝑦! = 𝑓(𝑥! ), they 𝑎𝑦! = 𝑓(𝑎𝑥! ), where 𝑎 = any constant.
2. If 𝑦! = 𝑓(𝑥! ) and 𝑦" = 𝑓(𝑥" ), then 𝑦! + 𝑦" = 𝑓(𝑥! + 𝑥" ).
Example: for the DC motor, applying 12 V, the angular velocity would be 1600 rpm. Next,
if we apply 6 V, the angular velocity would be 800 rpm. Then, the system obeys the first
superposition property, and the DC motor system is linear.
A physical system that demonstrates linearity (such as the DC motor) has a limited linear
range of operations. Therefore, we cannot increase the input voltage by a factor of 100 and
expect the corresponding angular velocity to increase by a factor of 100.

1.4 Modeling dynamic systems

• Developing an appropriate model is always the first step in system analysis.

• Mathematical models are obtained by applying the laws of physics to each element.
• The results obtained for a particular mathematical model are only approximated and are
valid only to the extent of the assumptions used to derive the model.
• The validity of a mathematical model can often by verified by comparing the model
solution (such as simulation results) with experimental results.
 Chapter 2 – Modeling Mechanical Systems

2.1 Introduction

This chapter introduces the fundamental techniques for deriving the modelling equations
for mechanical systems. These systems are composed of inertia, stiffness, and friction elements.
The mathematical models of mechanical systems are developed by applying Newton’s
laws of motion, which govern the interaction between force, mass, and acceleration.

2.2 Mechanical elements laws

A mechanical system is composed of inertia, stiffness, and energy-dissipation elements.

In addition, it may possess mechanical transformers, such as gears or levers.

Inertia elements
Inertia elements are either lumped masses (translational mechanical systems) or moment of
inertia (rotational mechanical systems). They are easily identified in Newton’s second law.
Force = mass x acceleration (translational system)
Torque = moment of inertial x angular acceleration (rotational system)
Therefore, the inertia element is the ratio of force and acceleration (or torque and angular
accelerations). A rigid body that has translational (“straight line”) motion has all its mass lumped
into a single element, 𝑚 (kg). A rigid body with purely rotational motion about an axis has all its
elements lumped into a moment of inertia, 𝐽 (kg-m2).
Inertial elements can store potential energy due to position in a gravitational field, or
kinetic energy due to motion. Potential energy 𝜉# of a mass m in a uniform field with gravitational
constant 𝑔 is.
𝜉# = 𝑚𝑔ℎ
where ℎ is the vertical position of the mass measured from a reference height. Kinetic energy 𝜉$
of mass m moving with a velocity 𝑥̇ is
1
𝜉$ = 𝑚𝑥̇ "
2
Kinetic energy of moment of inertia 𝐽 rotating with angular velocity 𝜃̇ is
1
𝜉$ = 𝐽𝜃̇ "
2
 The translational kinetic energy has dimensions of mass (𝑚) and velocity squared (𝑥̇ " ) or
units of kg-m2/s2, which is equivalent to N-m or joules. Also, the rotational kinetic energy has
dimensions of moment of inertia (𝐽) and angular velocity squared (𝜃̇ " ) or units of kg-m2 rad2/s2,
which is equivalent to N-m or joules.
All energy equation expressions must have the same units of N-m or joules.

Stiffness elements
When a mechanical element stores energy due to a deformation or change in shape, it
can be modeled as a stiffness element
Hook’s law states that the force required to stretch or compress a spring is proportional
to the displacement. Suppose a tensile force 𝐹 is applied to the free
end of a spring. The force required to produce displacement 𝑥 is
𝐹 = 𝑘𝑥
where 𝑘 is called the spring constant and has units of N/m
When both ends of a spring are free to move, then the force
required to stretch or compress a spring depends on the relative
displacement
𝐹 = 𝑘(𝑥" − 𝑥! )
A rotational mechanical system exhibits stiffness when a relationship
exists between an applied torque and the resulting angular displacement. A
simple example is the torsional shaft, where the left end of the shaft is fixed,
and the right end is free. An applied torque results in an angular
displacement, where the linear torque-displacement relation is
𝑇 = 𝑘𝜃
where 𝑘 is called the torsional spring constant and has units of N-m/rad. When both ends of the
shaft are free to rotate, the torque depends on the relative angular displacement
𝑇 = 𝑘(𝜃" − 𝜃! )
Stiffness elements can store potential energy due to deformations or deflections.
Potential energy 𝜉# stored by an ideal translational spring is
1
𝜉# = 𝑘∆𝑥 "
2
where ∆𝑥 = 𝑥" − 𝑥! , or the relative displacement between the free ends of the spring.
 Note that the units of energy in the previous equation is N-m, which is as the same units
for work (= force x displacement).
Potential energy of an ideal torsional spring is
1
𝜉# = 𝑘∆𝜃 "
2
where ∆𝜃 = 𝜃" − 𝜃! is the relative angular displacement between the free ends of the torsional
spring. Potential energy stored in a torsional spring has units of (N-m), which are again in the
same units as work.
When we model mechanical systems that possess stiffness elements, we simple lump all
of the system’s stiffness into a “spring element”.

Friction elements
When a mechanical element dissipates energy due to its
motion, it can be modeled as a friction element. Just as we used a
“spring element” to model stiffness in a mechanical system, we can
use a “damper” element to model friction. For a translational damper,
the damper is
𝐹 = 𝑏(𝑥̇ " − 𝑥̇ ! )
where 𝑏 is the viscous friction coefficient, with dimension of force/velocity, or units N-s/m.
Clearly, the damper force depends on the relative velocity between the piston/rod and cylinder.
Friction elements can only dissipate energy. The rate of energy dissipation from the damper is

𝜉%̇ = −(𝑏𝑥̇ )𝑥̇ = −𝑏𝑥̇ "

The previous equation shows that the rate of energy loss is always negative regardless of
the direction of the velocity. Power or energy time rate of change has units of N-m/s or J/s or
watts (W).
Next, consider a rotational or torsional damper, where the damping torque is
proportional to the relative angular velocity

𝑇 = 𝑏(𝜃̇" − 𝜃̇! )
Here b is the torsional viscous friction coefficient, with dimensions of torque/angular velocity, or
units of N-m-s/rad. An example of a rotational damper is a fluid drive, where two disks are
separated by a viscous fluid, and a friction torque exists if the relative angular velocity 𝜃̇" − 𝜃̇! is
not zero.
 When we model mechanical systems that involve energy dissipation due to friction, we
simply lump all of the system’s friction into a “damper element”, even though it may or not may
not be a physical piston-cylinder dashpot.

Mechanical transformers
Mechanical transformers are mechanical devices that transform an input motion of a
force. Common examples include levers and gear trains.
An ideal lever is rigid, has neither inertial nor friction,
and therefore cannot store or dissipate energy. Since the ideal
lever has no inertia, the moment about its pivot is zero, and
therefore 𝑓! 𝐿! cos 𝜃 = 𝑓" 𝐿" cos 𝜃; for small angles, cos 𝜃 ≈ 1
and 𝑓! 𝐿! = 𝑓" 𝐿" . If we consider force 𝑓! as the input force,
then the output lever force is 𝑓" = 𝑓! 𝐿! /𝐿" , which is greater than the input force when length
𝐿! > 𝐿" .
The other transformer is the gear
train, which may be used to increase or
decrease the angular velocity or torque
from the input axis to the output axis. In an
ideal gear train, the gears are assumed to
have zero inertia. The ratio of gear radii
(𝑟" /𝑟! ) is equal to the ratio of the number of gear teeth (𝑧" /𝑧! )
𝑟" 𝑧"
= = 𝐺. 𝑅
𝑟! 𝑧!
where 𝐺. 𝑅 is called the gear ratio. We can determine the angular velocity input-output ratio for
the gear train by
𝜔! 𝑟"
= = 𝐺. 𝑅
𝜔" 𝑟!
If the input axis gear has a smaller radius than the output axis gear, then 𝐺. 𝑅 > 1 and the
output velocity 𝜔" is less than the input angular velocity 𝜔! and the gear train is a speed reducer.
If 𝑟! > 𝑟" , then 𝜔" > 𝜔! and the output shaft rotates faster than in the input shaft.
 Since friction is neglected in an ideal gear train, the energy is transmitted without loss.
Work (or energy) for a rotational system is torque x angular displacement, and, therefore, power
is torque x angular velocity. Therefore, the power of the input shaft and output shaft yields
𝑇! 𝜔! = 𝑇" 𝜔" , and, therefore, the input-output torque ratio is
𝑇! 𝜔"
=
𝑇" 𝜔!
Therefore, when the gear train is a speed reducer, the output torque 𝑇" is greater than
the input torque 𝑇! .