1

State-space models 2
2nd order ODEs
J A Rossiter

Slides by Anthony Rossiter

 2

Introduction
• This resource focuses on derivations of state
space model equivalents for systems described
by ODEs.
• Here we consider 2nd order ODEs (see separate
resources for detailed derivation).
• The state-space model is defined in terms of the
derivatives of the states. If you know the
derivatives of all the states, then you can capture
the system behaviour.
• States relate to dynamic variables such as
displacement, height, tension, temperature, etc.
Slides by Anthony Rossiter
 3

States of a 2nd order ODE

1. The first step in forming a state space model is
to define the states.
2. There should be enough independent states to
capture the entire system dynamics – for low
order systems this selection is usually obvious.
3. However, when presented with a high order
differential equation, the user may have no
access (or information) relating to the definition
of the original underlying states, and thus an
arbitrary definition can be used.
Slides by Anthony Rossiter
 4

Modelling a mass-spring-damper
Force balance can be used to determine the overall
model of behaviour.
f 1  Bˆ v
f 2  kx
dv
f3 M
dt
f  f 1 f 2  f 3

dv ˆ
f M  Bv  kx States are v and x
dt
Slides by Anthony Rossiter
 5

States for mass-spring-damper

dv ˆ
f M  Bv  kx
dt
It is clear that the model contains states
velocity v and displacement x
so these are a logical choice.

For a state space model, find the derivatives of

each state and stack into a single vector.

Slides by Anthony Rossiter

 6

State-space model
Write the derivatives of v and x and stack in a
vector. dv dv f  kx  ˆv
B
M  f  kx  Bˆ v 
dt  dt M
dx dx
v v
dt dt

 dv  ˆ k In compact
 dt    B v  1 
   form
 dx   M M     M  f
   1   x  0 
0     z  Az  Bf
 dt   z B
A

Slides by Anthony Rossiter

 7

Resistor-inductor-capacitor in series
Consider the following circuit and use Kirchhoff’s
voltage law to derive an appropriate model.

dq
v1  iR1  R1
dt
di di dq  1 
v2  L
dt vL R   q
1 dt dt  C 
v3  q
C
v  v1  v 2  v3 States are i and q
Slides by Anthony Rossiter
 8

State-space model
Write the derivatives of i and q and stack in a
vector. q
di q v  iR
L  v   iR di C
dt C 
 dt L
dq
i di
dt q
dt
 di  In compact
 dt    R  1   i   1  form
 dq    L LC      L  v
   1  
0    0 
 q
 dt   z  z  Az  Bv
A B

Slides by Anthony Rossiter

 9

DC servo
The model can be
summarised with
some simple
equations.
Here it may be less
obvious which states
to choice due to the di
possibilities of two
v  L  kwL  iR
dt
angles/velocities,
ˆ dw
current, back emf and ki  BwL  J L
torque. dt
Slides by Anthony Rossiter
 10

DC servo choice of state

The states of interest must be defined with an
equation including their derivative.
Other states could be viewed as outputs (possible
measurements) and in this case will be linearly
dependent on the selected states.
Two obvious choices are wL and i:

di dw
v  L  kwL  iR; ki  Bˆ wL  J L
dt dt
Slides by Anthony Rossiter
 11

State-space model of DC servo

Express the derivatives of the selected states in a
column vector.
di ˆ dwL
v  L  kwL  iR; ki  BwL  J
dt dt
di v k R dwL k Bˆ
  wL  i ;  i  wL
dt L L L dt J J
 dwL   Bˆ k 
 dt   J   wL   0  z  Az  Bv
 di    k J 
    1 v
   R   i   
   L
 dt  
L  L  z B
Slides by Anthony Rossiter A
 12

DC servo with displacement

Should the user be interested in the angular
position as well as the angular velocity, then an
additional state is needed.
di ˆ dwL d L
v  L  kwL  iR; ki  BwL  J ; wL 
dt dt dt
 dwL   Bˆ k 
 dt   J 0
wL   0 
 di   J  
 k R    1 
    0  i    v
 dt   L L   L 
 L
d   1 0 0    
 L   0
 dt   

Slides by Anthony Rossiter 
 13

Pendulum
• A pendulum of length l
with end mass m is able to length - l
swing freely (assume some 
friction – constant k).
• A model can be derived mass - m
using force balance in the
tangential direction (small
angles).
ml  mg  kl

• Choose states:  ,   w
Slides by Anthony Rossiter
 14

Pendulum state space model

State derivatives: length - l
dw d 
ml  mg  klw w
dt dt mass - m

 dw 
 dt    kl  mg   w 0 
 d    ml ml       f
   1   
0     0
 dt    z B
A

Slides by Anthony Rossiter

z  Az  Bf
 15

Summary
Illustrated state-space model derivation for several
common 2nd order systems.
It is noted that the selection of states is important, but this
choice is often obvious from the underlying component
and balance equations as only certain states will have their
derivatives defined explicitly.
The compact form has vectors of states/inputs and
matrices of coefficients.

x  Ax  Bu
Slides by Anthony Rossiter
 Anthony Rossiter
Department of Automatic Control and
Systems Engineering
University of Sheffield
www.shef.ac.uk/acse

© 2016 University of Sheffield

This work is licensed under the Creative Commons Attribution 2.0 UK: England & Wales Licence. To view a copy of this licence, visit
http://creativecommons.org/licenses/by/2.0/uk/ or send a letter to: Creative Commons, 171 Second Street, Suite 300, San Francisco, California 94105, USA.

It should be noted that some of the materials contained within this resource are subject to third party rights and any copyright notices must remain with these materials
in the event of reuse or repurposing.

If there are third party images within the resource please do not remove or alter any of the copyright notices or website details shown below the image.

(Please list details of the third party rights contained within this work.

If you include your institutions logo on the cover please include reference to the fact that it is a trade mark and all copyright in that image is reserved.)