MOTION IN A STABILITY REGION (PART II)

CHAPTER 5
MOTION IN A STABILITY REGION (PART II)
Fig. 5 1
This section begins by showing how to find the steady-state response of a one
degree-of-freedom system acted on by a periodic excitation. The periodic
excitation has a period T, or equivaenty, a frequency
T

. The periodic
excitation is represented as a inear combination of harmonic functions. The inear
combination of harmonic functions is caed a Fourier series. "nce a periodic
excitation is expressed as a Fourier series, the steady-state response of a system
acted on by a periodic excitation is found. #y the principe of inear superposition,
the response of the system to the periodic excitation is a inear combination of the
responses of the harmonic functions that ma$e up the Fourier series.
%fter showing how to represent a periodic excitation by a Fourier series and how
to determine the associated response, it&s shown how to represent a periodic
excitation by a compex Fourier series. The compex Fourier series is used to
deveop a method of finding the steady-state response to a non-periodic excitation.
'xcitations are generay non-periodic. 'arthqua$es and wind produce non-
periodic excitations on buidings. (ind, road surfaces and guides produce non-
periodic excitations on vehice systems. )i$e the periodic excitation, the non-
periodic excitation is represented as a inear combination of harmonic functions.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
*owever, instead of its frequencies being mutipes of the frequency of an
excitation, there is a continuous range of frequencies. The non-periodic excitation
is an integra of harmonic functions instead of a discrete sum of them. The integra
of harmonic functions is caed the Fourier integra. The coefficients in the integra
are caed the frequency response. The frequency response represents the
ampitude of the response as a function of the frequency of the harmonics that
ma$e up the non-periodic excitation. +ince any non-periodic excitation can be
expressed in terms of its frequency response and, conversey, a frequency response
can be found for any non-periodic excitation, the two expressions are aso caed
the Fourier transform and the inverse Fourier transform, together they&re caed the
Fourier transform pair.
The Fourier transform pair is important in engineering. -t appies not ony to
excitations in vibrating systems but can be used to characteri.e how any physica
quantity changes in time. %fter the Fourier transform pair is deveoped, this section
deveops the procedure for finding the transient and steady-state responses of
systems acted on by non-periodic excitations. The procedure uses a variation of the
Fourier transform caed the )apace transform.
1. Fourier Serie
% periodic excitation f/t0 is represented as a Fourier series of sine functions and
cosine functions as
/5 10
T
r t C t B B t f
r
r
r r r r


!
, 0 sin cos /
!
1
0 /
1
1
+ +

in which Br and Cr are constants, caed Fourier coefficients, and T is the period of
the excitation, the frequency of the excitation is
T

. The periods
r
T
T
r
r

!
of the individua harmonics in 'q. /2 10 are integer fractions of the
period of the excitation. Therefore, the period of the series is T /+ee Fig. 5 10.
The Fourier coefficients are
1
.
1
The Fourier coefficients Br /r = 1, 1, 3 0 are found by mutipying both sides of
'q. /14.2 10 by

! 5
! 5
cos 0 /
T
T
r
tdt . +imiary, Cr /r = 1, !, 3 0 are found by
mutipying both sides of 'q. /5 10 by

! 5
! 5
sin 0 /
T
T
r
tdt . "n the right sides,
one then notices that

! 5
! 5
1 sin cos
T
T
s r
tdt t for any r and s and that

! 5
! 5
cos cos
T
T
s r
tdt t

! 5
! 5
1 sin sin
T
T
s r
tdt t
for any r and s that are
distinct
0. / s r

CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
/5 !0
... !, 1, sin 0 /
!
... !, 1, , 1 cos 0 /
!
! 5
! 5
! 5
! 5

r tdt t f
T
C
r tdt t f
T
B
T
T
r r
T
T
r r

%s an iustration, consider the square wave excitation shown in Fig. 5 !. -t is

defined over the interva 6-T5! T5!7 as

'

< <
< <

1 ! 5
! 5 1
0 /
1
1
t T A
T t A
t f
+ubstituting f/t0 into 'q. /5 !0

... !, 1, 0 cos 1 /
!
... !, 1, , 1 1
1
r r
r
A
C r B
r r

The Fourier series of the square wave excitation, truncated to n terms, is

T
t r
r
r
A
t f
n
r

!
sin 0 cos 1 /
1 !
0 /
1
1

Fig. 5 ! shows the Fourier series of the square wave truncated to n = ! terms and
to n 8 11 terms. The square wave itsef was actuay generated by truncating the
Fourier series to n = 511 terms.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
Fig. 5 !
MOTION IN A STABILITY REGION (PART II)
!. S"e#$%&S"#"e Re'o(e "o Perio$i) E*)i"#"io(
(hen the steady-state responses xr/t0 to individua excitations fr/t0 are $nown, then
the steady-state response to any inear combination of the excitations fr/t0 can be
found by the principe of inear superposition. )et a system be acted on by the
excitation


... 0 / 0 / 0 / 0 /
4 4 ! ! 1 1
+ + + t f A t f A t f A t f
The system&s steady-state response is simpy

... 0 / 0 / 0 / 0 /
4 4 ! ! 1 1
+ + + t x A t x A t x A t x
s

9otice that the coefficients in the inear combination of the excitation are the same
as the coefficients in the inear combination of the steady-state responses.
:onsider the damped system described by 'q. /5 !0. -ts steady-state response to a
constant force of F1 was found to be x = F15k. -ts steady-state response to
t F f cos
1

is x =
0 cos/
1
t X
and its steady-state response to
t F f sin
1

is x =
0 sin/
1
t X
. -t foows from the principe of inear
superposition that the steady-state response to the period excitation in 'q. /5 10 is

/5 40
[ ] [ ] ( )

+ +
+ +
1
]
1

1
1
1
1
1 1 1
07 sin/ 0 cos/ 6
!
1
0 sin/ 0 cos/
1
!
1
0 /
r
r r r r r r r
r
r r r r r r r r s
t C t B x
k
B
t x C t x B
k
B t x


.
in which from 'q. /2 210
/5 20
1
1
1
1
1
]
1

,
_

+
1
1
]
1

,
_

!
1
!
!
!
1
1
!
tan
7 ! 6 1
1 1
n
r
n
r
r
n
r
n
r
r
k
X

For exampe, oo$ again at the square wave excitation iustrated after 'q. /5 !0.
%ssume now that it acts on an undamped system that has mass m and stiffness k.
From 'q. /5 20 its steady-state response is

,
_

,
_

t
T
r
T
r
m k
r
r
A
t x
n
r
s

!
sin
!
cos 1 1 !
0 /
1
!
1
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
The response xs is shown in Fig. 5 4 for the Fourier series truncated to n = 1 term
and n = 11 terms. 9otice that xs is dominated by the first harmonic. The higher
harmonics contribute to the response ony sighty.
Fig. 5 4
+. Co,'-e* Fourier Serie
The compex Fourier series is deveoped beow as an intermediate step that wi be
used in the next section to represent a non-periodic excitation by harmonic
functions. (e begin by expressing the cosine and sine functions that appear in 'q.
/5 10 in terms of compex exponentias. +oving for the cosine and sine functions
in 'q. /2 ;0
/5 50
0 /
!
1
sin 0 /
!
1
cos
t i t i
r
t i t i
r
r r r r
e e
i
t e e t



+
+ubstituting 'q. /5 50 into 'q. /5 10, yieds the compex Fourier series
/5 ;0
T
r e F t f
r
r
t i
r
r

!
0 /

in which Fr are compex Fourier coefficients, given by
/5 <0
... !, 1, , 1 0 /
1 ! 5
! 5
t t

r dt e t f
T
F
T
T
t i
r
r

=eriodic excitations can be represented by rea Fourier series per 'q. /5 10 or by

compex Fourier series per 'q. /5 <0 athough the rea Fourier series is generay
preferred over the compex Fourier series.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
.. Fourier I("e/r#- (Fourier Tr#(0or,)
% non-periodic excitation can be thought of as a periodic function whose period T
is infinite. Foowing this reasoning, the non-periodic excitation is represented in
terms of harmonic functions by a Fourier integra obtained from the compex
Fourier series by etting its period T approach infinity. +ubstitute
/5 >0
r
F T F
T
0 /
!

into 'q. /5 ;0 and drop the index r to get

t i
r
r
t i
r
r r
e F T e F t f
!
1
0 /
%s T approaches infinity /and approaches .ero0 we get the Fourier integra of
f/t0
/5 ?0

d e F t f
t i
0 /
!
1
0 /

+imiary, substitute 'q. /5 >0 into 'q. /5 ;0, drop the index r, mutipy the
resut by T, and et T approach infinity to get the frequency response of f/t0
/5 110

dt e t f F
t i
0 / 0 /
'quation /5 ?0 can be thought of as transforming a frequency-domain function to
a time-domain function and 'q. /5 110 as doing the opposite. Thus F/0 is aso
caed the Fourier transform of f/t0 and 'q. /5 ?0 and /5 110 are caed a Fourier
transform pair.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
Fig. 5 2 @nit -mpuse
%s an exampe, the unit puse shown in Fig. 5 2 is

'

> <
< <

5 or 5 1
5 5 1
0 /
t t
t
t f
From 'q. /5 110, the frequency response of the puse is /+ee fig. 5 50

5 sin
! 0 / F
Fig. 5 5 frequency response of a @nit -mpuse
5. Di)re"e Fourier Tr#(0or,
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
Fourier transforms and inverse Fourier transforms are cacuated by integration.
-nstead, it can be simper to use a discrete Fourier transform in which the
cacuations are performed by sums. (hen using the discrete Fourier transform,
the excitation f/t0 is represented as samped data. The samped data is the
excitation evauated at the times t1, t1, t!, 3 tN-1 in which N is the number of
sampes. The samped data of f/t0 is f1, f1, f!, 3. fN-1. The samped data of f/t0 is
expressed in terms of compex exponentias as

/5 110 0 1 , ... ! , 1 , 1 / ,
1
1
1
!

,
_

N s e F
N
f
N
r
N
rs
i
r s

in which Fr /r = 1, 1, 3 N 10 are frequency data that indicates the amount of

,
_

N
rs
i
e
!
that is present in the samped data. The frequency data are aso caed
discrete Fourier coefficients. The discrete Fourier coefficients are
!

/5 1!0 0 1 , ... ! , 1 , 1 / ,
1
1
!

,
_

N r e f F
N
s
N
rs
i
s r

Figure 5 ; shows the compex harmonics

,
_

N
rs
i
e
!
/r, s = 1, 1, 3 , N 10 in the
compex pane. 9otice they represent a set of unit vectors that are equay spaced
around a circe.
!
The discrete Fourier coefficients Fr /r = 1, 1, 3 N 10 are found by mutipying
both sides of 'q. /14.5 20 by 0 1 ... , 1 , 1 /
1
1
!

,
_

N n e
N
s
N
ns
i

. "n the right
side, one then notices that

,
_

1
1
0 /
!
N
s
n r
N
s
i
e

is equa to .ero when
0 / n r
and is
equa to N when
0. / n r

CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
Fig. 5 ;
MOTION IN A STABILITY REGION (PART II)
%s an iustration, the samped puse shown in Fig. 5 < is

'

1 ,... 1 , 1
1 ,... 1 , 1
1
N n n s
n s A
f
s
From 'q. /5 1!0, its discrete Fourier coefficients are
0 1 , ... ! , 1 , 1 /
,
1
1
!
1

,
_

N r
e A F
n
s
N
rs
i
r

9otice that the discrete Fourier

coefficients are compex. The discrete
Fourier transform is used to anay.e
and even ater the frequency content
of samped data. +amped data is aso
caed a digita signa. The methods
and techniques of anay.ing and
atering samped data are caed
digita signa processing. Aigita
signa processing is used in many
engineering fieds. Aigita signa
processing is used to anay.e and ater
signas in acoustics, eectronics, and
imaging, to name a few.
1. L#'-#)e Tr#(0or,
The )apace transform is used in a genera procedure to sove constant-coefficient
inear differentia equations. The procedure draws from the method in :hapter 2
and the previous sections in this :hapter of finding transient soutions and steady-
state soutions to systems acted on by non-periodic excitations obtained using a
Fourier transform. The )apace transform of f/t0 is defined as
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
=art 1
=art !
=art 4
Fig. 5 ;
MOTION IN A STABILITY REGION (PART II)
/5 140

1
0 / 0 / dt e t f s F
t s
9otice that the )apace transform of f/t0 is simiar to the Fourier transform of f/t0
/+ee 'q. /5 1100. There are two differences, though. First, the ower imit in the
)apace transform is 1 instead of . This change was made because, when
computing time responses of systems, it&s convenient to et the excitation f and the
response x start at time t = 1. +econdy, notice that s is used in the )apace
transform instead of .
To deveop the )apace transform procedure in a genera way, we start with the n-
th-order constant-coefficient inear differentia equation
/5 120 f x a
dt
dx
a
dt
x d
a
dt
x d
a
dt
x d
a
n
n
n
n
n
n
+ + + +

1 1
!
!
!
1
1
1
...
The )apace transform is used to sove 'q. /5 120 in three steps.
/10 Ta$e the )apace transform of both sides of 'q. /5 120. This converts a
inear differentia equation into a inear agebraic equation in terms of
0 /s X
.
/!0 +ove for
0 /s X
and express it as a inear combination of the )apace
transforms that appear in the Tabe of )apace transforms.
/40 Ta$e the inverse )apace Transform of
0 /s X
using the Tabe of )apace
Transforms to obtain x/t0.
To perform the first step, we wi need to ta$e the )apace transform of the eft side
of 'q. /5 120 which contains derivatives of x/t0. From 'q. /5 140 and by
empoying integration by parts, the )apace transform of
0 /t x
is
1
1
1
1
0 1 /
0 / 0 / B 0 / 0 / 0 / x s X s dt e t x s e t x dt e t x s X
t s t s t s
+

+imiary, the )apace transform of

0 /t x
is

7 6 0 / 7 0 / 6
0 / B 0 / 0 / 0 /
1 1
!
1 1
1
1
1
0 1 /
v sx s X s v x s X s s
dt e t x s e t x dt e t x s X
t s t s t s

+

#y repeating this procedure for higher derivative of x/t0, the )apace transform of
the n-th derivative of x/t0 is expressed in terms of the )apace transform of x/t0 as
/5 150
1
1
]
1

+ + + +

1
1 1 !
!
1
!
1
1
1
0 /
... 0 / 0 / x s
dt
dx
s
dt
x d
s
dt
x d
s X s s X
n n
n
n
n
n
n n
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
'quation /5 150 is used to ta$e the )apace transform of the eft side of 'q. /5
120. The )apace transform of the right side of 'q. /5 120 is performed using the
Tabe of )apace transforms.
The second step in the )apace transform procedure is to sove for
0 /s X
. (hen
this is done, you wi find that
0 /s X
is a ratio of poynomias in s. The Tabe of
)apace transforms contains simpe ratios, certainy not every possibiity.
Therefore,
0 /s X
needs to be expressed in terms of the )apace transforms that
appear in the Tabe. This is done by performing a partia fraction expansion of
0 /s X
.

'xampes at the end of this section iustrate the )apace transform procedure.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
Tabe 1C )apace Transforms
-n cosing, this section deveoped methods of representing a non-periodic function
in terms of its harmonics and a procedure for finding the time response of an n-th
order constant-coefficient inear differentia equation. Figure 5 > shows the
notation that was used in this chapter.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
Function x/t0
)apace transform 0 /s X
n
n
dt
x d
1
1
]
1

+ + + +

1
1 1 !
!
1
!
1
1
1
... 0 / x s
dt
dx
s
dt
x d
s
dt
x d
s X s
n n
n
n
n
n
n
1
s
1
n
t
1
D
+ n
s
n
t
e

+ s
1
t n
e t

1
0 /
D
+
+
n
s
n

t cos
! !
+ s
s
t sin
! !

+ s
t
e t

cos
! !
0 /

+ +
+
s
s
t
e t

sin
! !
0 /

+ + s
Fig. 5 >
Di'-#)e,e(" For)e
"i,e re'o(e x/t0 f/t0
)o,'-e* Fourier Xr Fr
)oe00i)ie("
Fourier "r#(0or, X/0 F/0
L#'-#)e "r#(0or, 0 /s X
0 /s F
MOTION IN A STABILITY REGION (PART II)
E*#,'-e 1: Fourier I("e/r#- (Fourier Tr#(0or,)

Find and graph the frequency response of the puse shown. )et A1 8 1 and T = 5.
Solution

07 1 /cos 6sin 0 1 sin /cos 0 /
1 1
1
1
1
1
+

T i T
A
T i T
A
i e
i
A
dt e A F
T
t i
T
t i

The frequency response is compex. -t can be written in the poar form as

i
e F F 0 / 0 /
in which
! !
tan tan
!
cos
!
sin !
!
sin !
tan
sin
1 cos
tan
!
sin
!
!
cos 1
2 7 0 1 /cos 6sin 0 /
1
!
1 1
1
! 5 1
1 ! 5 1 ! ! 1
T T
T T
T
T
T
T A T A
T T
A
F

,
_

,
_

,
_

,
_

1
]
1

,
_

+

The magnitude of the frequency response is graphed in Fig. b. The magnitude of
the puse&s frequency response is maximum at = 1 and periodicay .ero.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH

Fig. 1a

Fig. 1b
MOTION IN A STABILITY REGION (PART II)
E*#,'-e !: Di)re"e Fourier Tr#(0or,

The cosine function t
T
A t f
!
cos 0 /
1
is samped N times at time intervas of
. 5 n T t Find the discrete Fourier transform of f/t0 for the two casesC 10 n = 15
and !0 n = 11. )et N = !11 and A1 8 1.
Solution
The discrete Fourier transform of the cosine function is
0 1 , ... ! , 1 , 1 / ,
!
cos
1
1
!
1

,
_

N r e
n
s
A F
N
s
N
rs
i
r

Figures c and d show its magnitude for n = 15 and n = 11. -n the first case, the
cosine function is samped 15 times per period. The discrete Fourier transform
shows a pea$ at N5n = 14.4 from the start and finish of the frequency ine. -n the
second case, the cosine function is samped 11 times per period. The discrete
Fourier transform shows a pea$ at N5n = 11 from the start and finish of the
frequency ine. -n the second case, the discrete Fourier transform is .ero except at
the two pea$s because N5n is an integer.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
Fig. !a
Fig. !c
Fig. !d
Fig. !b
MOTION IN A STABILITY REGION (PART II)
E*#,'-e +: L#'-#)e Tr#(0or,
% system is described by the differentia equation ; ; 5 + + x x x
with the initia
conditions x/10 8 1 and
. 1 0 1 / x
Find x/t0.
Solution
The )apace transform procedure and Tabe 1 are used. @sing the Tabe of )apace
transforms, begin by transforming the differentia equation
s
X s s
;
0 ; 5 /
!
+ + from which
0 ; 5 /
1
;
!
+ +

s s s
X
9ext, perform a partia fraction expansion of X . Factor the denominator into s, s+! and
s+4. Then rewrite the poynomia in terms of the )apace transforms that appear in the
tabe as

4 !
0 ; 5 /
1
!
+
+
+
+
+ +
s
C
s
B
s
A
s s s

in which A, B, and C need to be determined. 'quating powers of s in the numerators of
both sides yieds the three inear agebraic equations
1 ; C 1 1 ! 4 5 C 1 C
!
+ + + + A C B A s C B A s
Thus, A = 15;, B = -45;, and C = !5;. The )apace transform is

4
!
!
4 1
+

+

s s s
X
Finay, ta$ing the inverse )apace transform of X using the Tabe of )apace transforms,
the time response is

t t
e e x
4 !
! 4 1


The system is an over-damped. -ts equiibrium position is xe = 1. The answer given in 'q.
/e0 is chec$ed by differentiating x, substituting the resuts into the eft side of the
differentia equation, and verifying that it&s equa to the right side.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
E*#,'-e .: L#'-#)e Tr#(0or,
% system is described by the differentia equation t x x 4 cos 11 > ! +
in which x/10
8 1 and
. 1 0 1 / x
Find x/t0.
Solution
The )apace transform procedure and Tabe 1 are used. Transform the differentia
equation to get
! !
!
4
11 0 > ! /
+
+
s
s
X s
from which
0 4 0/ ! /
5
! ! ! !
+ +

s s
s
X
9ext, perform a partia fraction expansion of X . (rite
! ! ! ! ! ! ! !
4 ! 0 4 0/ ! / +
+
+
+
+

+ + s
D Cs
s
B As
s s
s
in which A, B, C, and D need to be determined. 'quating powers of s in the
numerators of both sides yieds
1 ? 2 C 1 1 ? 2 C 1 C 1 C
! 4
+ + + + D B C A s D B s C A s
This yieds A = 155, B = 1, C = 155, D = 1. The )apace transform is
! ! ! !
4 ! +
+
+

s
s
s
s
X
Ta$ing the inverse )apace transform of X using the Tabe of )apace transforms,
the time response is
t t x 4 cos ! cos +
9otice that the system is undamped.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
E*#,'-e 5: L#'-#)e Tr#(0or,
% free under-damped system is described by the differentia equation
1 + + kx x c x m with the initia conditions x/10 8 x1 and
. 0 1 /
1
v x
Find the time
response x/t0.
Solution
The )apace transform procedure is used. From the Tabe of )apace transforms,
the transform of the differentia equation is
1 0 / 0 /
1 1 1
!
+ + X k x X s c sx v X s m
from which
k cs ms
cx mv s mx
X
+ +
+ +

!
1 1 1
0 / 0 /
+ince the system is under-damped, the denominator can be written as
7 0 6/
! ! !
d
s m k cs ms + + + + in which m k
n
5 , m c ! 5 , and
! !

n d
. The poynomia is written in terms of the )apace transforms that
appear in the tabe as
k cs ms
cx mv s mx
X
+ +
+ +

!
1 1 1
0 / 0 /
8
! !
1 1 1
0 /
0 ! /
d
s
x v s x

+ +
+ +
8
! !
1 1
! !
1
0 /
0 /
0 /
d
d
d
d
s
x v
s
s
x

+ +
+
+
+ +
+
Ta$ing the inverse )apace transform of X using the Tabe of )apace transforms,
the time response
0 sin cos / sin cos
1 1
1
1 1
1
t
x v
t x e e t
x v
e t x x
d
d
d
t t
d
d
t
d

+
+
+
+


CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
E*#,'-e 1: L#'-#)e Tr#(0or,
Aerive the foowing )apace transformsC
10 L/10 8 15s, !0 L/
t
e

0 8
+ s
1

40 L/
t cos
0 8
! !
+ s
s
20 L/ t sin 0 8
! !

+ s
Solution
s
e
s
dt e L
st st
1 1
1 0 1 /
1 1

s
e
s
dt e e e L
t s st t t
1 1
0 /
1
0 /
1
! ! ! !
1
0 / 0 /
1
0 / 0 /
1
1 1
0 / 0 /
!
1 1 1
!
1
1 1
!
1
7 6
!
1
!
!
cos 0 /cos









+

+
+ +

1
]
1

+
+

1
]
1

+
+

+
+

s
s
s
i s i s
i s i s
e
i s
e
i s
dt e e dt e
e e
dt e
e e
dt e t t L
t i s t i s
t i s t i s st
t i t i
st
t i t i
st
! ! ! !
1
0 / 0 /
1
0 / 0 /
1
1 1
0 / 0 /
!
1 1 1
!
1
1 1
!
1
7 6
!
1
!
!
sin 0 /sin

+
+

1
]
1

1
]
1

s s
i s i s
i i s i s
e
i s
e
i s i
dt e e
i
dt e
i
e e
dt e
i
e e
dt e t t L
t i s t i s
t i s t i s st
t i t i
st
t i t i
st
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
Pro2-e, S"#"e,e("
The probems in this chapter consider systems 1 1 through 1- <. +et the appied
moment and the appied force to .ero. %ssume that the system is initiay at rest
and that the initia ange is 11
E
arger than the first equiibrium ange /i$e in
:hapter !0. This paces the system in the neighborhood of the first equiibrium
position.
Pro2-e, 5 3 1: Re'o(e 2% Fourier Serie
%ssume now that the oad is the square wave beow.
Fathematicay, the square wave can be written as

...0 , ! , 1 /
! 0 1 ! / ,
0 1 ! / 0 1 / ! ,
1 1 1
1 1 1
4

'

< <
< <
r
rT t T r A
T r t T r A
M
in which T1 denotes a haf-period. @se the same vaues of T1 and A1 here that you
used in =robem 2 1.
/a0 Find the response of the system by representing the forcing function M4 as
a Fourier +eries. =ot the response for about ; osciations.
Pro2-e, 5 3 !: Re'o(e 2% Su'eri,'oi(/ U(i" S"e' Fu()"io(
:onsider the oad described in =robem 5 1.
/a0 Find the response of the system by representing the forcing function M4 as
a series of step functions. %gain, pot the response for about ; osciations.
Pro2-e, 5 3 +: Re'o(e 2% Co(4o-u"io( I("e/r#-
:onsider the oad described in =robem 5 1.
/a0 Find the response of the system by representing the forcing function M4
using the convoution integra. %gain, pot the response for about ;
osciations.
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH
MOTION IN A STABILITY REGION (PART II)
CONTROL OF DYNAMICAL SYSTEMS: AN INTRODUCTORY APPROACH