Chapter 1

Introduction
to System
Dynamics
SAMANTHA RAMIREZ

1
Introduction

2
 What is System Dynamics?

The synthesis of
mathematical models to System Dynamics draws
represent dynamic on a variety of
responses of physical engineering specialties to
systems for the purpose of form a unified approach
analysis, design, and/or to study dynamic systems.
control.

Typically, building a prototype system and conducting experimental tests are either
infeasible or are too expensive for a preliminary design. Therefore, mathematical
modeling, analysis, and simulation of engineering systems aid the design process
immensely.

3
Challenge

https://www.youtube.com/watch?v=MLejkyXbJlc&feature=player_embedded

4
 When modeling a system
 What aspects of the system must you
consider?
 What tools, models, or information
will you need?

Challenge  How do you design or optimize the

system to ensure reasonable
Questions performance?
 What metrics do you use to measure
the system’s performance?
 How do you automate or control a
system?
 Where do you get started?

1) What is the system? Rider, bike, suspension (energy absorption), spring (store
energy), shock absorber (dissipates energy),
2) Answers to all questions are dependent on use of analysis.

5
 In this chapter you will: After completing this
chapter, you will be
Objectives come to a deeper
able to:
identify systems, subsystems,
& understanding of the art of
System Dynamics and the
and components,
identify potential applications
Outcomes purpose it serves in the
design, analysis, and control
of system dynamics in design
and analysis of mechanisms,
of physical systems, and and
begin to conceptualize how recognize and/or recall
a system is broken down into concepts used to represent
subsystems and components dynamic responses in other
to enable synthesis of engineering courses you are
mathematical models that or have previously taken.
represent the dynamics.

6
 Classification of Dynamic
Systems
Distributed System Lumped System
 Requires an infinite number  Involves a finite number of
of “internal” variables “internal” variables
 Variables are functions of  Variable are functions of
time and at least one time alone
spatial variable
 Governed by ordinary
 Governed by partial differential equations
differential equations (PDEs) (ODEs)

https://ccrma.stanford.edu/~jos/NumericalInt/Lumped_vs_Distributed_Systems.html

7
 Classification of Dynamic
Systems
Continuous-Time
Discrete-Time Systems
Systems
 Variables and functions  Variables defined only at
defined for all time discrete time points
 Similar to variables in the  Similar to variable in the
“analog” domain “digital” domain
 Described by differential  Described by difference
equations equations

http://signalsworld.blogspot.com/2009/11/continuoustime-and-discrete-time.html

8
 Classification of Dynamic
Systems

Time-Varying Systems Time-Invariant Systems

 System parameters vary with time  System parameters remain constant

http://slideplayer.com/slide/10398142/

9
 Classification of Dynamic
Systems
Linear System Non-linear System
 Obeys superposition  Does not obey
superposition
 Has homogeneity
 Does not have
homogeneity

https://study.com/academy/lesson/how-to-recognize-linear-functions-vs-non-linear-functions.html

Superpostition: f(x+y)=f(x)+f(y)
Homogeneity: f(ax)=af(x)

10
System Decomposition and Model
Complexity

A Quarter-Car Suspension
Model
 To formulate a model we must
identify the pertinent
components and formulate
mathematical representations
for each.
 The complexity of the model
depends on its intended use.

11
 Mathematical
Modeling and
Dynamic Systems
 Basic physical laws, theorems, etc. are
often modeled using mathematical
formulations.
 Newton’s Second Law
 Kirchhoff’s Equations
 In System Dynamics, differential and
algebraic equations are used to represent
dynamic responses.
 If they system of differential equations are
linear, we can take advantage of this to
use Linear Algebra or Laplace Transforms.

Inputs are variables that change the condition of the dynamic system
and can include things like external force, voltage sources, pressure
sources, etc.
Outputs are variables that are measured or observed to assess the
dynamic condition of the system
States are variables that are used to mathematically model the dynamic
behavior of the system.

12
 Analysis and Design of
Dynamic Systems

 We analyze systems to determine what

makes them function or respond as they do
so that we might be able to alter or optimize
their responses.
 Analyses are commonly conducted in the
time- or frequency-domains.
 Step responses usually entails time-domain
analysis
 Cyclic inputs entails frequency domain
analysis

Use a single input to determine the dynamic response.

Dynamic systems are often characterized in the time or frequency domain.

13
 Control of Dynamic Systems

 Control systems are used to automate dynamic responses;

that is to achieve the desired dynamic and static
characteristics.

 Automata or self-operating machines like the water clock

and fly-ball governor have existed for centuries.

Each block in a block diagram is a dynamic system.

14
 Diagrams of
Dynamic Systems
 A number of graphical
approaches are used to
represent or model dynamic
systems in a variety of
energy domains:
 Free-body diagrams
 Electric circuit diagrams
 Hydraulic circuit
diagrams
 Bond graphs are a
generalized graphical
method based on power
and energy.

15
  Bond graphs are a graphical approach
for diagramming the distribution and
flow of power and energy within a
dynamic system.
 Originally developed by Dr. Henry M.
A Graph- Paynter at MIT in 1959.
Centered  Bond graphing is a unified approach
that accounts for the storage,
Approach to dissipation, and conversion of energy
Modeling within a dynamic system.
 The bond graph accounts for the
input/output relations between
elements and subsystems of the
model that leads to computer
simulation of the dynamic response.

16
  Energy is defined as the capacity for doing
work.
 Power is defined as the rate of doing work
or the amount of energy consumed per
unit time.
Power and  Power (P) is also defined as the
Energy multiplication of an effort and a flow
 Effort (e): force-like variable
Variables  Flow, (f): velocity-like variable
 Effort and flow can be related to
generalized momentum (p) and
displacement (q), respectively.

Effort: Force (F), torque (τ), voltage (e), pressure (P)

Flow: linear velocity (v), angular velocity (ω), current (i), volume flow rate (Q)

17
 How is Energy Accounted For?

What can we do with

Tetrahedron of State
energy?

 Storage of Potential Energy

 Storage of Kinetic Energy
 Dissipation of Energy
 Transformation of Energy
 Energy Sources

Tetrahedron: 4-sided pyramid (5 sides: 4-triangular, 1-rectangular)

18
Effort, Flow, Momentum, and
Displacement Variables

19
 Momentum, Effort, Displacement, and
Flow

Generic Relationships for Effort and Flow

 Power 𝒫(𝑡) = 𝑒 𝑡 𝑓(𝑡)

 Generalized Momentum 𝑝 𝑡 = න 𝑒 𝑡 𝑑𝑡

𝑑𝑝
 Effort 𝑒 𝑡 = = 𝑝ሶ
𝑑𝑡

 Generalized Displacement 𝑞 𝑡 = න 𝑓 𝑡 𝑑𝑡

𝑑𝑞
 Flow 𝑓 𝑡 = = 𝑞ሶ
𝑑𝑡

Momentum: p=mv
𝑝ሶ = 𝑚𝑣
ሶ + 𝑚𝑣ሶ
𝑚𝑣ሶ goes to 0 if it isn’t losing mass
𝑚𝑣ሶ = 𝑚𝑎 = 𝐹
𝑚𝑎 = 𝑝ሶ = 𝐹

𝑝 = න 𝐹𝑑𝑡

Displacement: x
𝑥ሶ = 𝑣

න 𝑣𝑑𝑡 = 𝑥

20
Potential and Kinetic Energy

 Energy 𝐸 𝑡 = න 𝒫 𝑡 𝑑𝑡 = න 𝑒 𝑡 𝑓 𝑡 𝑑𝑡

𝑑𝑞
 Potential Energy 𝐸 𝑡 = න𝑒 𝑡 𝑑𝑡 = න 𝑒 𝑞 𝑑𝑞
𝑑𝑡

𝑑𝑝
 Kinetic Energy 𝐸 𝑡 =න 𝑓 𝑡 𝑑𝑡 = න 𝑓 𝑝 𝑑𝑝
𝑑𝑡

21
 Bonds, Ports, Signals, Inputs, and
Outputs

 Bonds connect elements at

power ports and represent an
effort-flow pair.
 Efforts or flows individually can
be represented by a signal in a
block diagram.
 Each element and port has an
input and output.
 Causal strokes are used in bond
graphs to indicate what end of
the bond has effort as in input.
 Port is a connection to
something else.

Effort on top or left. Flow on bottom or right.

Power direction is shown in bond graph not in block diagram.
Causal stroke shows where effort is going.
Port is a connection to something else. 1-port is shown in e, f.

22
 Word Bond Graphs

 A method for
decomposing a system by
identifying the more basic
components in words and
sketching the connections.
 Process
 Identify the basic
components of the system
 Connect interacting
components
 Identify the effort-flow
pairs

23
  Create a word bond
graph for the hydraulic
Word Bond Graphs actuator shown.

24
Challenge
Problem
CREATE A WORD BOND
GRAPH FOR THE SYSTEM
SHOWN THROUGH PART 7.

25
Summary

 System decomposition consists of breaking down the system into basic components
that can be readily characterized to enable modeling and mathematical
representation.

 Model complexity depends on use of the system representation and the necessary
accuracy of the predicted dynamic response.

 Mathematical models of dynamic systems commonly take the form of differential and
algebraic equations. As such, mathematical methods such as Linear Algebra and
Laplace Transforms are commonly used to analyze and design dynamic systems.

 Analysis is used to study dynamic systems and to characterize their responses. It can
be used to determine how changes in system parameters vary the dynamic response.

 When the desired dynamic response cannot be achieved through para- metric
optimization, automatic control systems can be employed to compensate and alter
the system response. Automatic controls are used to modify and/or automate
dynamic responses.

26
Summary Continued

 System decomposition consists of breaking down the system into basic

components that can be readily characterized to enable modeling and
mathematical representation.

 Model complexity depends on use of the system representation and the

necessary accuracy of the predicted dynamic response.

 Mathematical models of dynamic systems commonly take the form of differential

and algebraic equations. As such, mathematical methods such as Linear Algebra
and Laplace Transforms are commonly used to analyze and design dynamic
systems.

 Analysis is used to study dynamic systems and to characterize their responses. It

can be used to determine how changes in system parameters vary the dynamic
response.

 When the desired dynamic response cannot be achieved through para- metric
optimization, automatic control systems can be employed to compensate and
alter the system response. Automatic controls are used to modify and/or
automate dynamic responses.

27