100% found this document useful (1 vote)
334 views48 pagesChapter 4 Laplace Transform
Uploaded by
JELLIES -جليزCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
100% found this document useful (1 vote)
334 views48 pagesChapter 4 Laplace Transform
Uploaded by
JELLIES -جليزCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 48
nals
ai ey) EC eet
Resi ocala eiamen CHAPTER CONTENTS
as a spring/mass system or a
Say ae a 4.1 Definition of the Laplace Transform
either an external force f(é) or an 4.2 The Inverse Transform and Transforms of Derivatives
impressed voltage E(t). In 62:1 Inverse Transforms
oar een ie
problems in which fand E were 4.2.2 Transforms of Derivatives
continuous. However, 4.3 Translation Theorems
cea eae 4.3.1. Translation on the s-axis
have solved piecewise-linear 4.3.2 Translation on the t-axis
eee era ere ny 4.4 Additional Operational Properties
the Laplace transform discussed 4.4.1. Derivatives of Transforms
in this chapter is an especially 4.4.2 Transforms of Integrals
See eee 4.4.3. Transform of a Periodic Function
solution of such equations. . A
4.5 The Dirac Delta Function
4,6 Systems of Linear Differential Equations
Chapter 4 in Review 212
‘We wil assume thoughout
that's eal variable
»
Bl Definition of the Laplace Transform
INTRODUCTION In clementary calculus you learned that differentiation and integration are
‘ransforms—this means, roughly speaking, that these operations transform a function into another
function. For example, the function f(x) = x? i transformed, in (urn, into a linear function, a
family of cubic polynomial functions, and a constant by the operations of differentiation,
indefinite integration, and definite integration:
‘Moreover, these two transforms possess the linearity property; this means the transform
of a linear combination of functions is a linear combination of the transforms. For a and 2
constants,
a
Flas) + Bat] = afte) + Beta)
flare + Baw] dx afrinax ' Af ac ax
(de
and [en Bao] de af eae iB
provided each derivative and integral exists. In this section we will examine a special type of
integral transform called the Laplace transform. Inadition to possessing the linearity property,
the Laplace transform has many other interesting properties that make it very useful in solving
linear intal-value problems
If f(x, ) isa function of two variables, then a partial definite integral off with respect to
‘one ofthe variables leads toa function of the other variable. For example, by holding y con-
stant we see that ff 2xy* dx = 3y#. Similarly, a definite integral such as {2 K(s, O/dt
transforms a function f( into a function ofthe variables, We are particularly interested in
integral transforms of this last kind, where the interval of integration is the unbounded
interval (0,00).
I Basic Definition 11/(7)is defined for = 0, then the improper integral ff K(s, )/(dt
is defined asa limit
co = im ot 0
Ifthe limit exists, the integral is said to exist or to be convergent; ifthe limit does not exis, the
integral does not exist and is said to be divergent. The foregoing limit will, in general, exist for
only certain values of the variable s. The choice K(s, #) = e~ gives us an especially important
integral transform,
Peer eared
Let /be a function defined for 1 = 0. Then the integral
Lo} = { em fodt O)
is said to be the Laplace transform of f, provided the integral converges.
| CHAPTER 4 The Laplace Transform ‘The Laplace transform most likely was invented by Leonhard Euler, but is named after the
famous French astronomer and mathematician Pierre-Simon Marquis de Laplace (1749-1827)
who used the transform in his investigations of probability theory.
‘When the defining integral (2) converges, the result isa function of s. In general discussion, if
weuse a lowercase (uppercase) letter to denote the function being transformed, then the correspond-
ing uppercase (lowercase) letter will be used to denote its Laplace tansform, for example,
FSO} = Fis), Lg) = GS), Hy) = YE), and L(HED} = HK).
Using Defi
Evaluate (1)
SOLUTION From (2),
provided s > 0. In other words, when s > 0, the exponent ~sb is negative and e"* —+ 0 as
+b—>00. The integral diverges for s <0, =
‘The use of the limit sign becomes somewhat tedious, so we shall adopt the notation f? as
shorthand to writing lim,_,..(. j. For example,
erunar= =" =} soo.
At the upper limit it is understood we mean e+ 0 as £00 for s > 0.
EZNETEM Using Definition 41.1
Evaluate £ {1}
SOLUTION From Definition 4.1.1, we have (s} = Je" dt. Integrating by parts and
using lim. te“ = 0, sO, along with the reslt fom Example 1, we obtain
nape yp
“fea bamn 30) 3
mt!
y=
ECUTEM Using Definition 4.1.1
Evaluate (a) £(e") (by (0)
SOLUTION In cach case we use Definition 4.1.1
ran [oma
(a) He}
44 Definition ofthe Laplace Transform | 213 ‘The last result is valid for s > ~3 because in order to have lime“! =
that s+ 3>Oors> =3,
Owe must require
(by fer) = [eterna [e vat
© = 0 demands
In contrast to part (a), this result is valid for s > 6 because lim...
s—6>0ors>6.
BEC Using Definition
Evaluate {sin 21)
SOLUTION From Definition 4.1.1 and integration by parts we have
‘esi 29} = | esin 20d =
lim e7" cos 21
> 0 Laplace transform of sin 2r
+
P{ sin 21
‘AL this point we have an equation with (sin 2r) on both sides of the equality. Solving for
that quantity yields the result,
{sin 24}
s>0.
+4
Isa Linear Transform Fora sum of functions, we can write
[eterno + peintar= af'erpind + Bf erent
whenever both integrals converge for s > e. Hence it follows that
Haf@) + Ba) = aL FO) + BL(aO) = aFO) + BOC). @
Because of the property given in (3), is said to be a linear transform. Furthermore, by the
properties ofthe definite integral, the transform of any inte lineae combination of Functions fi),
fad). 0s the Sum of the transforms provided each transform exists on some common
interval on the sans.
ECE _ Linearity of the Laplace Transform
In this example we use the results of the preceding examples to illustrate the linearity of the
Laplace transform.
(a) From Examples | and 2.we know that both 2{1} and {1} exist on the interval defined
bys > 0, Hence, for s > Owe can write
a+ sy=90y +509 <1 3
214 | CHAPTER 4 The Laplace Transform (b) From Example 3 we saw that ‘£{e} exists on the interval defined by s > 6, and in
Example we saw that {sin 2r} exists on the interval defined by s > 0. Thus both transforms
exist for the common values of s defined by > 6, and we can write
tae" ~ s5sinny = 2 ten) —15s¢sin2 = 2 -
FIGURE 41.1 Piecowise-continuous (©) From Examples 1,2, and 3 we have for s > 0,
function
{Ue ~ 51+ 8} = 10SE{e*) ~ SL(1) + BEI}
5.8
‘We state the generalization of some of the preceding examples by means of the next theo-
rem, From this point on we shall also refrain from stating any restrictions on s; itis understood
that ¢ is sufficiently restricted to guarantee the convergence of the appropriate Laplace
transform,
FIGURE 41.2 Function fis of exponential
order (Re
Amor extensive Hist fp @ 2-4
transforms given in 7
— © 20) 123. © £e=—
@ Lisin ke © Fleosk) = Fs
(f) {sind ke) (g) L{cosh ks) =
st) oes ,
let I Sufficient Conditions for Existence of £ {f(t)} The integral that defines the
Laplace transform does not have to converge. For example, neither ‘£{1/t) nor £{e*} exists.
Sufficient conditions guaranteeing the existence of ‘P(0) are that be piecewise continuous
A ‘on [0, co) and that fbe of exponential order for 1 > 7. Recall that a function f'is piecewise con-
t————— tinuous on (0, co) if, in any interval defined by 0 = a =f = b, there are at most a finite number
‘of points k= 1,2,..-51 (fy © 4). at which fhas finite discontinuities and is continuous on
Oo) ‘each open interval defined by f,_., << 4, See FIGURE 41.1. The concept of exponential order
i) is defined in the following manner
Definition 412 Exponential Order
4 A function fis said to be of exponential order if there exist constants c, M > 0, and T> 0
such that /(91 = Me for all>.
»
If Fis an increasing function, then the condition \f()= Me", > T, simply states that the
fol graph of fon the interval (7, oo) does not grow faster than the graph of the exponential function
ae Me", where cis positive constant, See FIGURE 41.2. The functions f®) = tf) = e“, and
eos F(®) = 2 cos tare all of exponential order c = 1 for t > O since we have, respectively,
Ise, le I2eos r= 2e.
‘A comparison of the graphs on the interval (0, oc) is given in FIGURE 413.
‘A positive integral power of rs always of exponential order since, for ¢> 0,
°
FIGURE 41.3 Functions with blue graphs weisme® or (Slsa torr> 7
ane of exponential order
44 Definition of the Laplace Transform | 215 sro]
FIGURE 41.4 /(0) = e” isnotof
exponential order
,
2 —
a
FIGURE 415 Piccewise-continuous
function in Example 6
is equivalent to showing that lim, ../e" is finite for n = 1, 2,3, .. The result follows by
‘applications of L’Hopital’s rule. A function such as f(2) = eis not of exponential order since,
as shown in FIGURE 4.14, e” grows faster than any positive linear power of ¢ for > ¢ > 0. This
can also be seen from
a gia
e950 a8 1900
for any value of e. By the same reasoning, ¢¥e" oo as t+ 00 for any s and so the improper
integral f'e“e" at diverges. n other words, ‘P{e*) does not exist.
Pee ee
1 f() is piecewise continuous on the interval 0, oo) and of exponential order, then ‘E{ f())
exists fors > c.
PROOF: By the additive interval property of definite integrals,
x noy= [erroa+ [etna =n +h
“Te integral exists because ten be writen as asum of integral over intervals on which €~*/0)
is continuous. Now fis of exponential order, so there exists constants c,f > 0, T>> Oso that
(l= Me for > T. We ean then write
we Prima ealereu nfo
fors> Since JF Me“°-% dr converges, the integral / le~"f() dt converges by the comparison
test for improper integrals. This, in turn, implies that, exists for s> c. The existence of J, and
implies that PLO} = J5 e~°F(0) de exists for > c
EXE _ Transform of a Piecewise-Continuous Function
0, 0=1<3
Bratwte #179} orsin= {9 =F 3
SOLUTION ‘This piccewise-continuous function appears in FIGURE 415, Since fis defined in
two pieces, f(/(0)} is expressed asthe sum of two integrals:
£100} = { ef dt = [oa + { e"Oydt
REMARKS
‘Throughout this chapter we shall be concerned primarily wit functions tha ar both piecewise
contingous and of exponential order We note however, tha these two conltions are sufficient
but not necessary for the existence of a Laplace transform. The function f(t) = 17!? is not
piecewise continuous on the interval (0, o2); nevertheless its Laplace transform exists. Soe
Problem 43 in Exercises 4.1.
216 | CHAPTER4 The Laplace Transform QS BGG _ Answers to selected odd-numbered problems begin on page ANS-8
In Problems 1-18, use Definition 4.1.1 t0 find £{ (0)
2 po)
i“ 2)
FIGURE 4.1.8 Graph for FIGURE 41.7 Graph for
Problem 7 Problem 8
9. fo 10. fo)
1 a
\—. ++
FIGURE 4.1.9 Graph for
FIGURE 41.8 Graph for
Problem 9 Problem 10
1 fe? w f=’
13. f= te M f= Pe
15. fi) = eTsinr 16. fl) =e cost
17, fle) = teost 18. f= esine
In Problems 19-36, use Theorem 4.1.1 to find £(f(0)
B. f() = 40° ~ 5sin3r
3B. fd = sinh kt
3. fo =e sinh
flo) = €08 51+ sin 24
0) = cosh kt
0) =e" cosh
18. f= 24° a. fin =F
2. fi) = ar 2 f= 43
BFP 46-3 24. fi) = —4F + 166+
Bf) => 1" 2. f()
2. f)=14+ 2. fi) =P -¢
2 f=0+e% 0. fi) =(e- P
2.
u
38
In Problems 37-40, find SE(/() by first using an appropriate
twigonometric identity
31. f(t) = sin 2rc0s 2
38, (0) = sin(ar + 5)
38. fit) = cos?
4. fld) = 1Ocos(t — 716)
41, One definition of the gamma function Ia) is given by the
improper integral
r@ = [eo ‘edt, a> 0.
Use this definition to show that Pa + 1) = aP(a).
4 Use Problem 41 to show that
a+
Hey
‘This result isa generalization of Theorem 4.1.1(b).
In Problems 43-46, us the results in Problems 41 and 42 and
the fact that P'(@) = V7" to find the Laplace transform of the
given function,
fer? wm pose?
a5 f= 8 48 fo) = 60 — 240?
cussion Problems
F(f(0} = FAs) for s> cx, When does
LU + HO} = Fis) + Fats)?
48, Figure 4.1.4 suggests, but does not prove, that the function
{f(t) = e" isnot of exponential order. How does the observation
that ? >In M + ct, for M> O and r sufficiently large, show
that e” > Me for any ¢?
48, Use part (c) of Theorem 4.1.1 to show that
te +001 ib
eS a
where a and bare real and = —1, Show how Buler’s formula
(page 121) can then be used to deduce the results
Het eosb = Se
‘Pe sin by} = —2
a Hes OTE
50, Under what conditions is a linear function f(x) = mx + b,
im # 0, a linear transform?
51. The function f(t) ~ 2te"cose* is not of exponential order.
Nevertheless, show thatthe Laplace transform ‘F{2te"cose"}
exists (Hin Use integration by parts]
52, Explain why the function
1 o=r<2
= 44, zee
Ye- 4),
is not piecewise continuous on [0, 09).
441 Definition of the Laplace Transform | 217 58, Show that the function f(®) = 1/#? does not possess a
Laplace transform. [Hinr: Write F{1/} as two improper
integrals:
wry = [as
Show that, diverges.]
54, If £{/(} = Fls) and a > Ois a constant, show that
eye ~ L(2)
‘This result is known as the change of scale theorem.
In Problems 55-58, use the given Laplace transform and the
result in Problem 54 to find the indicated Laplace transform,
‘Assume that a and kare positive constants,
a ve) = LHe
& Hicont) = Ee Heats
Br PAe — sind} = Sa lhe ~ sink
58. Pfoost sit} = = 2
‘P{coske sinhkt}
218
42| The Inverse Transform and Transforms of Derivatives
INTRODUCTION In this section we take a few small steps into an investigation of how the
Laplace transform can be used to solve certain types of equations. After we discuss the concept of
the inverse Laplace transform and examine the transforms of derivatives we then use the Laplace
transform to solve some simple ordinary differential equations.
4.2.1 Inverse Transforms
[The Inverse Problem If F(s) represents the Laplace transform of a function f(0),
that is, £{f(0) = FO), we then say /() isthe inverse Laplace transform of F(s) and write
‘S(O = F-"(F()}. For example, from Examples 1,2, and 3 in Section 4.1 we have, respectively,
meg eG) = emt
‘The analogue of Theorem 4.1.1 forthe inverse transform is presented next,
pont! Pee
@
w ras{ A} nara. © e-2{ =
@ sou=s-{ 4} (©) coskr= 9 {3}
(© sonar {4} @ cosntr= 1
‘When evaluating inverse transforms, it often happens that a funetion of s under consideration
does not match exactly the form of Laplace transform F(s) given ina table. Ttmay be necessary
10 “fix up” the function of s by multiplying and dividing by an appropriate constant.
EXT Applying Theorem
a} @ ets}
Evaluate (a) {s
SOLUTION (a) ‘Tomatch the form given in part (b) of Theorem 42.1, we identify n + 1 =
‘orn = 4 and then multiply and divide by
(3) de
CHAPTER 4 The Laplace Transform Pasa ties: dine
Toei dma
(b) To match the form given in par (f) of Theorem 4.2.1, we identify = 7 and sok = V7.
‘We fix up the expression by multiplying and dividing by V7
a}
[£7 Isa Linear Transform The inverse Laplace transform is also a linear transform;
thats, for eonstants and B,
£
LaF) + BG) F(s)) + BENG), o
where F and G are the transforms of some functions fand g. Like (3) of Section 4.1, (1) extends
to any finite linear combination of Laplace transforms.
EXTEN _Termwise Division and Linearity
pranae {28}
SOLUTION We first rewrite the given function of sas two expressions by means of termwise
division and then use (1)
termwise division Linearity and fixing up constants
+
Sr ea} ieiae
= ~2e0s 21 + 3sin 2s, «parts (e) and (8) of Theorem 42.1 with k= 2
I Partial Fractions Partial fractions play an important role in finding inverse Laplace
transforms. As mentioned in Section 2.2 the decomposition ofa rational expression into eom-
ponent fractions ean be done quickly by means of a single command on most computer algebra
systems, Indeed, some CASs have packages that implement Laplace transform and inverse
Laplace transform commands. But for those of you without access to such software, we will
review in this and subsequent sections some of the basic algebra in he important cases in which
the denominator of a Laplace transform F(s) contains distinct linear factors, repeated linear
factors, and quadratic polynomials with no real factors, We shall examine each of these cases
4 this chapter develops
ETE Partial Fractions and Linearity
-f i+ 6s+9 i
G-De- D675
SOLUTION There exist unique constants A, B, and C such that
Evaluate
St os+9
C06 -D65 9D
_ Als = 2s + 4) + Bes = Ds + 4) + ls = 1s = 2)
= D6 DE+H
v2
Since the denominators are identical, the numerators are identical:
POF 9= Ay — e+ 4) +B — DE +N +CE—DE-D.
By comparing coefficients of powers of s on both sides of the equality, we know that (3) is
equivalent to a system of three equations in the three unknowns A, B, and C. However, recall
that there i a shortcut for determining these unknowns, If we set s = I, = 2, and s= —4
42 The Inverse Transform and Transforms of Derivatives | 219 in 3) we obtain, respectively,*
16 = A(- 116
BAY6), 1 = C(-5X-6),
Bc
+65+9 16S, 256 130
G=6- DEH Sat y=2 see
and sod
4. Hence the partial fraction decomposition is
‘and thus, from the linearity of
rat +6549 \
= D6- DEF.
and part (c) of Theorem 4.2.1,
4.2.2 Transforms of Derivatives
(4)
I Transform of a Derivative As pointed out in the introduction to this chapter, our im:
‘mediate goal is to use the Laplace transform to solve differential equations. To that end we need
to evaluate quantities such as SE(dyidt} and {d?y/dr?). For example, iff" is continuous for
1 0, then integration by parts gives
aro Perod-e4ol +s [enna
=f0) + s£(F0)
o LUPO) = sk) — FO)
‘Here we have assumed that e~" f(1) + 0 as t—+ oo. Similarly, with the aid of (6),
LAF) = [ emp" dt = ef")
==) + s#1F0)
= s{sF(s) — FO) —f'O) from (6)
or EUSO) = PFS) — sf(O) - f°).
if + [ erode
In like manner it ean be shown that
LPO} = SFE)
A) - sf O- FO)
(6)
i}
(8)
‘The recursive nature of the Laplace transform of the derivatives ofa function f should be appar-
cent from the results in (6), (7), and (8). The next theorem gives the Laplace transform of the nth
derivative of f
fre foie)
Uff"... are continuous on (0,00) and are of exponen
continuous on [0, ce), then
LLP) = FF) — SFO) — FO) == FO,
where Fis) = F(/(0)
“The numbers 1, 2, and ~4 are the zero ofthe common denominator (¢— 1K ~ 23
220 | CHAPTER4 The Laplace Transform
onder andit/"(N is piecewise
4. I Solving Linear ODEs Iris apparent from the general result given in Theorem 4.2.2 that
L{d"ylde"} depends on ¥(s) = E{y(2)) and the n ~ 1 derivatives of y(t) evaluated at r = 0. This
property makes the Laplace transform ideally suited for solving linear initial-value problems in
‘which the differential equation has constant coefficients. Such a differential equation is simply
‘linear combination of terms y, 9',"..... 9"
+ doy = 80s
2990) = oats
where the coefficients a,,=0,1,....m and yop Vie. +.Yq-1 are constants, By the linearity property,
the Laplace transform of this linear combination i a linear combination of Laplace transforms:
ay of tcy ey) <4
{3} * {o + ag L(y} = £{8(0}. @
From Theorem 4.2.2, (9) becomes
a\s'V) = P10) = = YO]
(10)
# ayaa M8) = 8200) = OI] + + a¥s) = Gs,
where £{y() = Y(s) and J
(9(0} = G(s). In other words:
The Laplace transform ofa linear differential equation with constant coefficients becomes
an algebraic equation in Ys)
If we solve the general transformed equation (10) for the symbol ¥(s), we first obtain P(s)¥(s) =
Q(s) + G(s), and then write
26) , Ge)
Hi) = 22, GO mn
© Be * Fe) m
where Ps) = ag + a6" 4+ ay (isa polynomial ins of degree less than oF equal
ton ~ 1 consisting ofthe various products ofthe cocticiens a, = 1,...m,and the prescribed
inital conditions yy, --»Yn-vatd GG) isthe Laplace transform of g(.* Typically we putthe
{wo terms in (11) over the Teast common denominator and then decompose the expression into
{wo oF more partial fractions. Finally, the solution (0) ofthe original initial-value problem is
340 = L(Y}, where the inverse transform is done term by term.
“The procedure is summarized in FIGURE 42.
Find unknown “Tansfomed DE
‘ye tha sates] iy apace wantoen ip of Bevomes an
DE andinisal | > APM Laplace arson * —*) cai equation
onions in¥@)
Solution oF ps __[ Soe tanstomnay
DE Lappy inverse wanton 2-1 ae
FIGURE 42.1 Steps in solving an IVP by the Laplace transform
‘The next example illustrates the foregoing method of solving DES.
ECT Solving a First-Order IVP.
Use the Laplace transform to solve the initial-value problem
1Bsin2, (0) =6.
‘The polynomial P(s) isthe same asthe mds degree auxiliary polynomial in (13) in Section 3.3, withthe
‘usual symbol m replaced by s
42 The Inverse Transform and Transforms of Derivatives. | 221 222
SOLUTION We first take the transform of each member ofthe differential equation:
2f® gy g
{2} cay = 2tsnan a
But from (6), £{dyidt) = s¥(s) — y(0) = 5¥(s) ~ 6, and from part (d) of Theorem 4.1.1,
{sin 2r} = 24(5? + 4), and so (12) is the same as
26
SY) = 6 + 3Y(9) = or (+ 3)N) $64
Solving the last equation for Y(s), we get
6, 26 65"
S43 CFD) CDE
Ms)
(13)
a
Parva factions: Since the quadratic polynomial s? + 4 does not factor using real numbers, its assumed
usin petypomst ‘numerator in the partial fraction decomposition is a linear polynomial in s:
6s? +50 Bot
CFDCFH S43 FHS
Putting the right side of the equality over a common denominator and equating numerators
gives 6s? +50 = ACs + 4) + Bs + Cs + 3). Setting s = —3 then yields immediately
A= 8. Since the denominator has no more real zeros, we equate the coefficients of s*and s:
6 =A + Band 0 = 3B + C. Using the value ofA in the frst equation gives B = —2, and then
using ths lat value in the second equation gives C = 6, Thus
6st +50
Yo) = 0
6+ +H
‘We are not quite finished because the last rational expression still has to be written as two
fractions. Buc this was done by termuise division in Example 2. From (2) of that example,
ras} fete} elo}
It follows from parts (c), (4), and (e) of Theorem 4.2.1 that the solution ofthe initial-value
problem is y() = 8e™™ = Deos2r + 3sin2r =
v=
(EXTEN Solving a Second-Order IVP
Solve y'-3y'42y=e%, YO=T, YO=S
SOLUTION Proceeding as in Example 4, we transform the DE by taking the sum of the
transforms of each term, use (6) and (7), use the given initial conditions, use part (c) of
‘Theorem 4.1.1, and then solve for Ys):
22} ~ xf carp ne
#Ys) ~ s9(0) = yO) ~ 3IS¥68) — MO] + 29) =
sea
(8 = 38+ DY 2+
uaa
st2 1
24 6549
342° @-e+DETD C-DE-DE=D
¥) (14)
and so y(t) = ‘£-'[Y(9)] The details of the decomposition of Y(s) in (14 into partial fractions
have already been carried out in Example 3. In view of (4) and (5) the solution ofthe intil-
value problem is y() = eel + Be" + hen“
CHAPTER 4 The Laplace Transform Examples 4 and 5 illustrate the basie procedure for using the Laplace transform to solve
4 linear initial-value problem, but these examples may appear to demonstrate a method that
is not much better than the approach to such problems outlined in Sections 2.3 and 3.3-3.6.
Don’t draw any negative conclusions from the two examples. Yes, there is a lot of algebra
inherent in the use of the Laplace transform, but observe that we do not have to use variation
of parameters or worry about the cases and algebra in the method of undetermined coeffi-
cients. Moreover, sinee the method incorporates the prescribed initial conditions direetly into
the solution, there is no need for the separate operations of applying the initial conditions to
the general solution y = eyy, + ex¥ + “+ + e,35c+ yp Of the DE to find specific constants in
‘a particular solution of the IVP,
The Laplace transform has many operational properties. We will examine some of these
properties and illustrate how they enable us to solve problems of greater complexity inthe sec-
tions that follow.
‘We conclude this section with @ litle bit of additional theory related to the types of
functions of s that we will generally be working with, The next theorem indicates that not
every arbitrary funetion of s is a Laplace transform of a piecewise-continuous function of
exponential order,
Theorem 423 Behavior of F(s) as 5
If fis piecewise continuous on (0, 00) and of exponential order, then lim Sf {F(}
PROOF: Since f(t)is piecewise continuous on the closed interval [0,11 itis necessarily bounded
‘on the interval. That is, [f( = M, = Me™. Also, because fis assumed to be of exponential
‘order, there exist constants , M, > 0, and "> 0, such that l/(0)l = Mze"" for ¢> T. ITM denotes
the maximum of (M,, Mf} and c denotes the maximum of {0, y}, then
werent | eyo ar =m
ee dt = — ME
for s > ¢. As s—.00, we have |%£{/(0)}| +0, and so ‘£( f()} 0. =
‘As a consequence of Theorem 4.2.3 we can say that functions of s such as F\(s) = 1 and
FAs) = si(s + 1)are not the Laplace transforms of piecewise-continuous functions of exponential
‘order since Fs) + 0 and F,(s) #0. as s ->c0, But you should not conclude from this that F,(s)
and F,(3) are not Laplace transforms. There are other kinds of functions
REMARKS
(H The inverse Laplace transform of a function F(s) may not be unique; in other words,
itis possible that £{ f(} = L{F,(0)} and yet f, +f. For our purposes this is not any-
thing to be concerned about. Iff, and f; are piecewise continuous on [0, co) and of ex
ponential order, then f, and f; are essentially the same. See Problem 50 in Exercises 4.2,
However, iff, and f, are continuous on [0, 00) and #£{ f,()} = £{ f,(2)), then f, = fz on
the interval.
(i This remark is for those of you who will be required to do partial fraction decompo-
sitions by hand, There is another way of determining the coefficients in a partial fraction
decomposition in the special case when £{ f(*)} = F(s) is @ rational function of sand the
denominator of F is a product of distinct linear factors. Let us illustrate by reexamining
Example 3. Suppose we multiply both sides of the assumed decomposition
B c
sea
Fto+9 A
@- De - D679
(15)
42 The Inverse Transform and Transforms of Derivatives | 223 224
by, say, s ~ 1, simplify, and then sets ~ 1. Since the coefficients of B and C on the right side
of the equality are zero, we get
Ft 6r+9
©2644)
Written another way,
‘where we have colored or covered up the factor that canceled when the left side was multiplied
bys ~ 1. Now to obiain B and C we simply evaluate the left-hand side of (15) while covering
vp, in turn, s~ 2 and s + 4:
2+ 649 25 246 +9 1
TH 6 G
‘The desired decomposition (15) is given in (4). This special technique for determining coef-
ficients is naturally known as the cover-up method.
ii) In this remark we continue our introduction to the terminology of dynamical systems.
Because of (9) and (10) the Laplace transform is well adapted to linear dynamical systems. In
(LD the polynomial P(s) = a,s" + 4,3"! + -+ + ag is the total coefficient of ¥(s) in (10)
and is simply the left-hand side of the DE with the derivatives d'yldt* replaced by powers s,
k= 0, 1, ...»n. Ibis usual practice to call the reciprocal of P(s), namely, W(s) = 1/P(s), the
transfer funetion of the system and write (11) as
Ys) = WS}Q) + WGC). (16)
In this manner we have separated, in an additive sense, the effects on the response that are duc
(o the intial conditions (that is, W(s)Q(s)) and to the input function g (that is, W(s)G(s)). See
(13) and (14), Hence the response (0) of the system is a superposition of two responses
0 = EOE) + LUMEIGO} = yal + 100.
Ifthe input is g(@) = 0, then the solution of the problem is yo() = £-(W(s)Q(s)}. This
solution is called the zero-input response of the system. On the other hand, the function
30) = £"{W(s)G(s)} is the output due to the input ¢(). Now ifthe initial state ofthe system
is the zero state (all the initial conditions are zero), then Q(s) =O, and so the only solution of
the initial-value problem is y,(0). The latter solution is called the zero-state response of the
system. Both y() and y,(0 are particular solutions: (isa solution of the IVP consisting of
the associated homogeneous equation with the given initial conditions, and y\(®) isa solution of
the IVP consisting ofthe nonhomogeneous equation with zer0 initial conditions. In Example S,
we see from (14) that the transfer function is W(s) = 1/(s? — 35 + 2), the zero-input response is
gi f_st2
wy ea}
and the zero-state response is
n3e! + de%,
yO =
“f 1 Lagden
— Ft dt de,
© De-D6+ Hf ~ 5°76
Verify that the sum of yy(2) and y,(0) is the solution y() in that example and that (0)
_y§(0) = 5, whereas y\(0) = 0, y4(0)
TS
CHAPTER 4 The Laplace Transform FER BGS _Answers to selected odd-numbered problems begin on page ANS-2
4.21 Inverse Transforms
422 _Transtorms of Derivatives
In Problems 1-30, use Theorem 4.2.1 to find the given inverse
transform,
&
5-3
2 {3 __}
(s = V3ys + V3)
BON 6- aaa}
nae! ft}
5e—DGE+ De
ats a aaa}
ne!
aaa * “es
2 oa) * ea)
In Problems 31 and 32, find the given inverse Laplace transform,
by finding the Laplace transform ofthe indicated function f.
2 eras) f= enh in
af 4). s ~ cosbr — cova
@. {= + ays? + aH fo m
In Problems 33~4, use the Laplace transform to solve the given
initial-value problem.
a®
1
ar
|, M0 =0
2245-0, 90)
mw 224y=0, v0)
By 6c, yO=2
6. y" = 2eos St, y(0) =
37. y' + Sy'+4y=0, WO)= 1, yO) =
38. y" — dy’ = be" —3e% 0) = 1, YO =
3. yf ty = VEsin Vix, (0) = 10. yO) = 0
@. y+ 9y=e', (0) = 0, yO) =
M, 2" + 3y —3y' —2y= eK) =O, YO) =0, =I
Wy" + Dy" =p! = y= sin3e, y0)=0, YO) =0, YO=1
1. y= y=, 0) = 0,y") = 0,9") = 0," = 2
My = 9) = 0, yO = 1y'O = 0,y") = 3," =0
In Problems 45 and 46, use one of the inverse Laplace transforms
found in Problems 31 and 32 to solve the given initial-value
problem,
45. y" + dy = 10cos5t, (0) = 0,9")
4, y" + 2y = sins, (0) = 0,y'Q =0
‘The inverse forms ofthe results in Problem 49 in Exercises 4,1 are
tr
{eave} =
aarF
In Problems 47 and 48, use the Laplace transform and these
inverses to solve the given initial-value problem,
Wy ty =e cos 2, (0) =
4 y= 2y' + 5y=0, KO =1,y'O=3
and £
Discussion Problems
4, (a) With a slight change in notation the transform in (6) is
the same as
LAF OY = 5 LF) ~ FO).
With f(0) ~ re, discuss how this result in conjunction
with part (c) of Theorem 4.1.1 can be used to evaluate
Lhe
(b) Proceed as in part (a), but this time discuss how to use
(with (9 = ¢sin krin conjunction with parts (4) and (e)
of Theorem 4.1.1 to evaluate (1 sin kt}
42 The Inverse Transform and Transforms of Derivatives | 225 50, Make up two functions f, and fs that have the same Laplace 82 Suppose f(t) is a function for which "(0 is piecewise con-
transform, Do not think profound thoughts
51, Reread Remark (ii) on page 224, Find the zero-input and the
zero-state response for the IVP in Problem 38.
FIGURE 43.1 Shifton saxis
226
‘inuous and of exponential order c, Use results in this section
and Section 4.1 to justify
$10) = lim SF,
where F(s) = Othe graph of F(s) is shifted aunts to the right, whereas if <0,
the graph i shifted lal units to the left, See FIGURE 43,
For emphasis tis sometimes useful to use the symbolism
FO) = LF Over
Where ss — a means that in the Laplace transform F(s) of (9 we replace the symbol s where
it appears by s — a.
Using the First Translation Theorem
Evaluate (a) £(e"?) — (b) £le* cos 4}
SOLUTION The results follow from Theorems 4.1.1 and 4.3.
3!
@) He) = LO pues
Jars ©
a?
FEB hone CF HIG
I Inverse Form of Theorem 4.3.1 To compute the inverse of F(s — a) we must recognize
Fs) find f() by taking the inverse Laplace ransform of Fs), and then multiply f() by the expo-
nena funetion This procedure ean be summarized symbolically inthe following manner:
LFS = a) = FF aread = FO, a
where fl) = £-F))
(b) Le cos 41) = F608 48) pin
CHAPTER 4 The Laplace Transform ESZOTEM Partial Fractions and Completing the Square
motes pate {28} g of
Tinea factors in (3.
oy
SOLUTION (a) A repeated linear factor isa term (s ~ a)", where a is a real number and nis
‘a positive integer = 2, Recall that if (5 ~ a)" appears in the denominator of a rational expres-
sion, then the assumed decomposition contains m partial fractions with constant numerators
and denominators s ~ a, (§ ~ a), ..,(8 — a)". Hence with a = 3 and n = 2.we write
ast
oy SH
oe
‘The numerator obtained by putting the two terms on the right over a common denominator is
2s +5 = A(s— 3) + B, and this identity yields A = 2 and B = 11. Therefore,
ats 2 hn
cy 53°C a
a orthotics} 8
Now I/(s — 3)°is F(s) = 1/5? shifted 3 units to the right. Since S£~!{1/s"} = 1, it follows from
peat wf pen fe] peer
(by To start, observe that the quadratic polynomial s* + 4s + 6 does not have real zeros and
so has no real linear factors. In this situation we complete the square
v4 5B S458
s (+2 6)
‘Our goal here i to recognize the expression on the right as some Laplace transform F(s) in
which s has been replaced throughout by 5 + 2. What we are trying to do here is analogous
to working part (b) of Example I backwatd, The denominator in (5) is already in the correct.
form; that is, #2 + 2 with s replaced by s + 2. However, we must fix up the numerator by
manipulating the constants: $s + $= Ks +2)+$—3= He +2) +3
‘Now by termwise division, the linearity of £7, parts (d) and (¢) of Theorem 4.2.1, and
finaly from (1),
CFD ye +42
sot ora}
(aL ara &
Me sin Vi me
+58 _URs+H+2B 1 s42 2 1
3
1 x
Se cos Vir +
5 1:
Ini falue Problem.
6) F9y= Fe = 2, OA IT
SOLUTION Before transforming the DE note that ts right-hand sides similar the fonction
in part (a) of Example 1, We use Theorem 4.3.1, the initial conditions simplify, and then
Solve
43 Translation Theorems | 227 solve for 9) = LUO}
Ly") ~ 6L{y'} + 9F(y) = L(re*)
SY) = s¥(0) = yO = 6[8Y%(8) = HO] + 9%
oy
(8 = 65 + (9) = 25 +5
(= YG) = 2545+
245 2
XG —
“ 6-3 6-3
"The first teem on the right has already been decomposed into individual partial fractions in
(2) in part (a) of Example 2:
¥)
8)
rm so ae
From the inverse form (1) of Theorem 4.3.1, the last two terms are
wb foes aw
and so (8) is y() = 2e* + Lire + are™
EE A initial-value Problem
Sole "+4 F6y=1+e% O=0 YO=O.
SOLUTION Ly") + AL (y') + 6L(y) = LU) + He)
> Lt
#119) = 590) = 'O) + ALM) = yO] + 619) = FF
4 ds oy) = 221
@ +404 0N) = 2A
yoy =_—5* 1 _
36> DE FET O
Since the quadratic term in the denominator does not factor into real linear factors, the partial
fraction decomposition for Y(s) is found to be
6 1B sR +5
MOT Peas
‘Moreover, in preparation for taking the inverse transform, we have already manipulated the
last term into the necessary form in part (b) of Example 2. So in view of the results in (6) and
v2
. :
wan) de tch) detatah stele
cos Vir ~ V2 6->in Vit :
228 | CHAPTER4 The Laplace Transform 8
FIGURE 432 Graph of unit step function
y
FIGURE 433 Function can be writen
Fa) = = 9UG=1)
fe)
2}
a
1 ui
FIGURE 43.4 Function can be writen
Jig) = 2=3UG=2) + U3)
FIGURE 435 Function in Example S
4.3.2 Translation on the taxis
IL Unit Step Function Inengincering, one frequently encounters functions that are either
‘off” or“on.” For example, an external force acting on a mechanical system or voltage impressed
‘on acircuit can be turned off after a period of time. Its convenient, then, to define a special func-
tion thats the number 0 (off up toa certain time = a and then the number | (on) after that time.
‘This function is called the unit step function or the Heaviside function, named after the renowned
English electrical engineer, physicist, and mathematician Oliver Heaviside (1850-1925). The
Heaviside layer in the ionosphere, which can reflect radio waves, is named in his honor.
De Ee ine
‘The unit step function ‘(1 ~ a) is defined to be
0, O=r 0 the
raph ofthe function y = /(@ ~ a) U0 ~ a coincides with the graph of y = f(t — a) fort = @
VW (which s the entre graph of y = f() = 0, shifted a unis to the right onthe axis) but siden
telly zero for0 = ta
| _, "We saw in Theorem 43.1 that an exponential mule of f() resus in a translation ofthe
transform FG) onthe vans, Asa consequence ofthe next theorem we see tat whenever PU) is
@ prea multiplied by an exponential unetion &~, a > Othe inverse transform of the product e““F3)
vo) isthe uncon fshfted along the Fanis inthe manner stated in Figure 4.3.60) This rest
Presented next in its direct transform version, sealed the second transation theorem of
{ second shifting theorem.
1 ___.
IL Fis) = £{f()} and a > 0, then
{fe — ayU(r— a)} = e™ Fi).
fe) 4-2)
FIGURE 436 Shift on-axis
PROOF: By the additive interval property of integrals, fi?e™"f(t ~ a) U(s— a) dican be written
as two integrals:
visa aua-ay= | esa auc ade [ e"fe— a) UC - a) dt
* KH a
revo ferds1 0, is
Ee FI} = flt— aU ~ a. (15)
230 | CHAPTER4 The Laplace Transform ECE _ Using Formula (15)
wate er} gy fee)
SOLUTION (a) With the identifications a = 2, Fis) = Ms ~ 4), £-(F(s)} = e, we have
from (15)
aa = etiBayer = 2),
(b) With a = 2/2, F(s) = si(6? + 9), £-UF(9)} = c08 34, (15) yields
‘The last expression can be simplified somewhat using the addition formula for the cosine
Verify thatthe result is the same as —sin 3°11 (¢ — m2) =
I Alternative Form of Theorem 4.3.2 Weare frequently confronted with the problem
‘of finding the Laplace transform of a product ofa funetion g and a unit step function "U(t — a)
where the function g lack the precise shifted form f(t — a in Theorem 4.3.2. To find the Laplace
transform of (Ct ~ a, itis possible to fixup g() int the required form flt — a) by algebraic
‘manipulations, For example, if we want to use Theorem 4.3.2 to find the Laplace transform of
PALE — 2), we would have to force a() = #? into the form f(s ~ 2). You should work through
the details and verify that = (¢~ 2)° + 4(¢ — 2) + 4 is an ieniy, Therefore,
#(rUG= 2)
((@— 2° UE 2) + Ae 2)UG— 2) + UE 2},
2% de
by Theorem 4.3.2. But since these manipulations are time-consuming and often not obvious, it
is simplerto devise an alternative version of Theorem 4.3.2. Using Definition 4.1.1, the definition
of ‘(7 — a), and the substitution u = 1 ~ a, we obtain
Beacrue~ on = | e%eCoa [ew atu + a) du
Thats, L(g UE— a} = e-*E{Qlt + a). (16)
With a = 2 and g(t) = # we have by (16),
LEP UE = Dy = e™*H + 2}
= eM E(P + ar + 4}
44
+3+5)
‘which agrees with our previous calculation
(EXTEN _ Second Translation Theorem—Alternative Form
Evaluate ‘£{cos #4U(¢ — =).
SOLUTION. With g(0) = cos, a= =, then g(t + =)
formula for the cosine function, Hence by (16),
cos (1 + m) = —cos ry the addition
L{cos 1U(t— m)} = ~e-**£ {C08 4}
43 Translation Theorems | 231 In the next two examples, we solve in turn an initial-value problem and a boundary-value
problem, Both problems involve a piecewise-linear differential equation,
An Value Problem
sone 498An, xO"s, whew f= [,, OST
SOLUTION The function f can be written as f(?) ‘cos t4li(t — 7) and so by linearity, the
20714 it =the em
s¥(s) — y(0) + Y(s)
Ys) = i)
[Now proceeding as we did in Example 7, it follows from (15) with a = # that the inverses of
the terms in the bracket are
5
wage m Hf oe} esge nen
eat
“{
an PHA oor = cost = my UE -
5 {yer}
anf
In Problems 21-30, use the Laplace transform to solve the given
initial-value problem,
a, eo, 0) =2
2. yO) =0
2. 1 = 1, ¥O
2. Pe, 0) =0, yO =0
3. 4 9 =0, YO=1
3. y"— 4 H4y=P, YO=1, yYO=0
2. y"— 6" + 13y=0, (0) =0, y= -3
2B. 2y" + 20y' + S1y=0, MO =2, y'O)=0
2B. yy =ecost, =O, YO
B. y=2y FSY=1 FH O)=0,
In Problems 31 and 32, use the Laplace transform and the
procedure outlined in Example 10 to solve the given
yO=4
boundary-value problem.
By Hy Fy=0, ¥O=2, ya) =2
@ y"+8y' + 20y=0, 10) =0, ym =0
33. A 4-th weight stretches a spring 2 ft. The weight is released
from rest 18 in above the equilibrium position, and the
resulting motion takes place in a medium offering a damp-
ing force numerically equal to } times the instantancous
velocity. Use the Laplace transform to find the equation of,
motion x(0),
‘34, Recall that the differential equation for the instantancous
charge q(0) on the capacitor in an LRC-series circuit is
aq dg 1
Sh RE + a= BO. (20)
234 | CHAPTER 4 The Laplace Transform
See Section 3.8. Use the Laplace transform to find q() when
L=Th,R= 200, C= 0.0056, EQ) = 1S0V,1> 0,410) = 0,
and i) = 0. What isthe eureent i)?
35, Consider the battery of constant voltage E> that charges the
capacitor shown in FIGURE 6.310, Divide equation (20) by L
anddefine 2A = RiL and. = VLC. Use the Laplace transform
{o show that the solution q(t) of q’ + 2Aq’ + wg = El,
subject to q(0) = 0, (0) = 0,8
aft =e4(oadV
FIGURE 43.10 Circuit in Problem 35
36, Use the Laplace transform to find the charge q() in an
RC-series when 4(0) = Oand E(t) = Be™™, k> 0. Consider
two cases: k # WRC and k = URC.
4.32 Translation on the t-axis
In Problems 37-48, find either Fs) or (9, as indicated.
m. La-YUE-D} 3% BEAU}
3. £02) @ Z[Gr+ NMED)
M1. Leos UU — a) Lf sinsUCE— ID}
4
on{t + =}
oe}
ees} «etal
In Problems 49-S4, match the given graph with one of the given
Functions in (a)-(f). The graph of f(s given in FIGURE 43.11,
fe
B
i
FIGURE 43:11 Graph for Problems 49-54 @ fo - fo Ue a
() fie - 5) UE = B)
© fOUE-a)
@ fo ~ FO UE
© fe Ue- a) ~ fH) UE)
©) fle a) Ue ~ a) ~ fle a) Ue B)
8)
‘ 10) {\
4 , oe
2 i
FIGURE 43:12 Graph for FIGURE 43.13 Graph for
Problem 49 Problem 50
1. 0)
ttt.
FIGURE 42.18 Graph for
Problem SI
2 fo
+t
mrs
FIGURE 4315 Graph for
Problem 52
53. fo 5. so)
+++ - ,
oP i
FIGURE 43.18 Graph for FIGURE 43.17 Graph for
Problem 53 Problem 54
In Problems 55-62, write each function in terms of unit step
functions. Find the Laplace transform of the given function,
2 osr<3
13
re
40
sing 1 = 32
sin, O05 1< 20
a so={* see
a fo)
1.
T
rectangular pul staeease funtion
FIGURE 43.18 Graph for FIGURE 43.19 Graph for
Problem 61 Problem 62
{In Problems 63-70, use the Laplace transform to solve the given
initial-value problem,
osc)
63. y Fy =f, (0) =O, where fl) = { at
se yy, 101-0, ae £0 {
4
0,
8. YF 4y =F, yO) = 0, yO) = ~1, where
wv ears 901-0, tae = {
1 osrei
10 TS
gy 18) sinstne 29, no yo) <0
68 y"— Sy" F6y=WE-1), yO) =O, y"O=1
8. y+ y= fi, WO) = 0, yO) = 1, where
0, 0
SO = SM St Qe
0, 12 le
1, y' + dy’ + 3y=1- UE—2)— UE 4) + UE 6),
yO) = 0, »°(0)
TN, Suppose a mass weighing 32 Ib stretches a spring 2 ft If
the weight released from rest at the equilibrium position,
find the equation of motion x(¢) if an impressed force
F() = 201 acts on the system for 0 = ¢-< 5 and is then
removed (see Example 5). Ignore any damping forces. Use
4 graphing utility to obtain the graph x(#) on the interval
(0, 10}
72. Solve Problem 71 if the impressed force f(#) = sin racts on
the system for 0 = < 2zr and is then removed.
In Problems 73 and 74, use the Laplace transform to find the
charge q(0) on the capacitor in an RC-series circuit subject to the
ziven conditions.
7, gO) =0,R=250,
C= 0.08 £, E( given
in FIGURE 4.320
1, q(0)=quR=100,C=01f,
F(0) given in FIGURE 43.21
43 Translation Theorems | 235 FIGURE 4320 £0) in
Problem 73
FIGURE 4321. £(0) in
Problem 74
(a) Use the Laplace transform ( find the current i(0) in a
single-loop LR-series cireuit when (0) = 0, L = 1h,
R= 10.0, and E() isas given in FIGURE 4322.
Use a computer graphing program to graph i(1)
for 0-5 #5 6, Use the graph 10 estirsate fg 8d gin the
maximum and minimum values of the current,
respectively.
)
FIGURE 43.22 £(0 in Problem 75
(a) Use the Laplace transform to find the charge 4(#) on the
capacitorin an RC-series circuit when q(0) = 0, = S00,
C= 0.01 f, and E( is as given in FIGURE 4323.
(b) Assume Fy = 100'V, Use a computer graphing program
tograph q() for 0=#= 6, Use the graph to estimate dru,
the maximum valve ofthe charge.
Fo
FIGURE 4323 £0) in Problem 76
‘A cantilever beam is embedded at its left end and free at its
right end. Use the Laplace transform to find the deflection
x) when the load is given by
{i o [cos (A ~ B) ~ cos (A +B)
If wesetA = kr and B= k(¢~ 7), we can carry out the integration in (6):
{sta} 1 | teostear = 1) — cost] dr
enter 9 =r]
2
‘Multiplying both sides by 24? gives the inverse form of (5).
L Transform of an Integral When g(1) = 1 and $#(9(9)} = G(3) = 1s, the convolution
‘theorem implies that the Laplace transform of the integral of fis
a
‘The inverse form of (7),
)
CHAPTER 4 The Laplace Transform FIGURE 442 LRC-series circuit
‘can be used in lieu of partial fractions when s"is a factor of the denominator and f(#) = £~"{FUs)}
is easy to integrate, For example, we know for f(t) = sin# that Fis) = Ms? + 1), and so by (8)
{tg} < (M22) « [Sorte
{to} - fae DY = [0 -cosndr = 1 sine
- L fur + DY fi clase
ofc} = aE) [ie oncnden P= 1 cs
‘and so on,
I Volterra Integral Equation The convolution theorem and the result in (7) are useful
in solving other types of equations in which an unknown function appears under an integral sign.
Tn the next example, we solve a Volterra integral equation for 0),
LO = a+ { Syne = 9 ae. )
‘The functions g(#) and (2) are known, Notice that the integral in (9) has the convolution form
(2) with the symbol A playing the par of g
ECE An Integral Equation
Solve fl) = 3? - t=
[sovemar for fo.
SOLUTION nthe integral we identity 4(¢ — +) = e*"* so that h(t) = e'. We take the Laplace
transform of each term; in particular, by Theorem 4.4.2 the transform of the integral is the
product of £{f()} = Fis) and £(e") = Ws — 1):
FQ) = 3-5
FO)
After solving the last equation for F(s) and carrying out the partial fraction decomposition,
we find
Fe)
1
sce
‘The inverse transform then gives,
no = 30{2
BP a+ 1 en =
I Series Circuits Ima single-loop or series circuit, Kirchhof?’s second law states that the
sum of the voltage drops across an inductor, resistor, and capacitor is equal to the impressed
voltage E(0). Now it is known thatthe voltage drops across an inductor, resistor, and capacitor
are, respectively,
di 1
= Ri, and =| ite)dr,
pe Ri, chi
‘where i) is the current and L, R, and C are constants, It follows that the current in an LRC-series
circuit, such as that shown in FIGURE 442, is governed by the integrodifferential equation
1
di *
Lt RU = GI iar
©. (10)
44 Additional Operational Properties | 241 FIGURE 443 Graph of current) in (11)
of Exampl
242
os
le?
zs
EXETER An ntegroditferential Equation
Determine the current i() in a single-loop LRC-circuit when L = 0.18, R= 20,C= 0.18,
(0) = 0, and the impressed voltage is
(= 1201 = 120044 = 1).
SOLUTION Using the given data, equation (10) becomes
018 +214 10f eerar= t200- 120rK¢ - 1)
Now by (1. £(Jpile dr) = Hf, where 0) = $i). Thus the Laplace transform ofthe
integrodiferenial equation is
ho
ony an © 022 ] -mabstsaes
“Multiplying this equation by 10s, using s? + 20s + 100 = (s + 10)? and then solving for (3)
ives
5 ¥ 107
a
By partial fractions,
1100, 0. 1 |
+ 0 dt _
S10" "FIO we 10
From the inverse form ofthe second translation theorem, (15) of Section 4.3, we finally obtain
fo) = 12[1 — WE 1)] -12[e" ~ @- 0)
= 120167" = 1o80¢ = He“ PUG =D,
‘Written as a piecewise-defined function, the current is
pp (12 ~ Be = 1201e—% <1
KO Vaaren = 126! = 120%e-M = Logo = ew,
my
‘Using the last form of the solution and @ CAS, we graph s(t) on each of the two intervals and
then combine the graphs. Note in FIGURE 443 that even though the input E(t) is discontinuous,
the output or response i) isa continuous function, =
I Post Script—Green’s Function Redux By applying the Laplace transform to the
initial-value problem
Yi tay’ + by=FO, yO) =0,y'O) = 0,
‘where a and b are constants, we find that the transform of y(0) is
Fo)
% =a
where F(s) = £( (0). By sewriting the foregoing transform as the product
Yo)
FO)
CHAPTER 4 The Laplace Transform we can use the inverse form of the convolution theorem (4) to write the solution of the IVP as
{
(0) of Section 3.10 thatthe solution of the IVP is also given by
x) — n)flr)dr, (12)
where = g(@) and I1(F(9)} = f(). On the other hand, we know from
v= [ou nnerdr, (13)
‘where G(r, 7) isthe Green's function forthe differential equation.
‘By comparing (12) and (13) we see that the Green’s function for the differential equation is
}
GU.) = at 7). (1a)
For example, forthe initial-value problem y" + 4y = f(0), (0) = 0, »"(O) = Owe find
JnFuample $0 Seaton 310, > Thus from (14) we se thatthe Green's Function for the D
setcicgmeecauiee P Tesag ) SocEuamplet in Seton 310
By’ + dy =f is Gl.) = gt- 71
REMARKS
Although the defining integral in the Laplace transform was known and used for along time
prior to the twentieth century it was not used to solve differential equations. The fact that we
Use the Laplace transform today to solve a variety of equations is actually due to Oliver
Heaviside (see page 229). In 1893, Heaviside invented an operational calculus for solving
differential equations encountered in electrical engineering, Heaviside was no mathematician,
and his procedures for solving differential equations were formal manipulations or procedures
Jacking mathematical justification, Nonetheless, these procedures worked. In an attempt t0
put is operational calculus on a sound foundation, mathematicians discovered thatthe rules
of his calculus matched many properties ofthe Luplace transform. Over time, Heaviside’s
operation calculus disappeared, to be replaced by the theory and applications of the Laplace
transform,
‘You should verify either by substitution in the equation or by the method of Section 2.3
that y(0) = ¢"'fie"du is a perfectly good solution of the linear initial-value problem
y' + y= e', x0) = 0. We now solve the same equation with a formal application of the
Laplace transform. If we denote £{y}= HS) and Z{e"} = Fis), then the transform of the
equation is
99) +H) =F oe iy = FO
1
mi
Using £—"{F(6} = ef and {
convolution theorem that the solution of the initial-value problem is
x) w{ros +} [e tMdr = e'
With « playing the part of u, ths is the solution as first given. Do you see anything strange in
this solution method of the IVP?
44 Additional Operational Properties | 243 4.4.3 Transform of a Periodic Function
I Periodic Function 1faperiodic function fhas period 7, 7 > 0, then f(r + T) = f(t). The
Laplace transform of aperiodic function can be obtained by integration over one period.
Doreen eed
Iff(@is piecewise continuous on [0, oc), of exponential order, and periodic with period 7, then
LEGO} =[ efi0 dt
PROOF: Write the Laplace transform of fas two integrals;
gio) = { esoa + | epai
Yo} = | eo at [ “flo dt
When we let = u + 7 the last integral becomes
evpdt = [ ee My + T)du = [eon =
‘Therefore PFO [ero de> TF f(O).
Solving the equation in the ast line for S£{/(#)} proves the theorem, =
bo (EEZUETEINY_ Transform ofa Periodic Function
ee Find the Laplace transform of the ptiodic function shown in FIGURE 444
+tiHi il, SOLUTION The function £() is called a square wave and has period T= 2, For 0 = t
1203 4 E(t) can be defined by
FIGURE 444 Square wave in Example 8
and outside the interval by f(t + 2) = f(). Now from Theorem 4.4.3,
‘nna = all
sasenae4
er (15) =
AP d Volta
‘The differential equation forthe current i() i a single-loop LR-series circuit is
toe R= Bo. (16)
Lo + Ri = BU
Determine the current i(t) when (0) = Oand E(t) the square-wave function given in Figure 4.4.4.
SOLUTION Using the result in (15) of the preceding example, the Laplace transform of the
DEis
Iss) + Ris) =——— or ts) =
s(t em s(s + RIL) ™
244 | CHAPTER 4 The Laplace Transform a a
FIGURE 4.45 Graph of current) in
Example 9
| 4.4 | Exercises MN
‘To find the inverse Laplace transform of the last function, we first make use of geometric
series. With the identification x = e~, s> 0, the geometric series
1 1
Sl -xdata xt to becomes se1-
Thx Tre
1 UR _uR
From —_-=
5@+RD 5
wwe can then rewrite (17) as
ZC Loaner se
R\s se)" Fe"
vt )- espace ea 5+ RL )
By applying the form of the second translation theorem to each term of both series we
obtain
= 2(t-aw- 9 ra) —a0-3 +)
este prema 2) emma -3)+ 0)
or, equivalently,
1 ity 4 WS ea = eM —MQCUE = a)
f= Rd eM) + PCV yU(E Ay.
‘To interpret the solution, Iet us assume for the sake of illustration that R = 1, L = 1, and
0=1<4,Inthis case
yae-D+d-e
i =1-e"-a-
in other words,
44.1 _ Derivatives of Transfo
In Problems 1-8, use Theorem 44
Laplace transform.
1 £ (te)
3 L{re0s 20)
5 £{rsinh}
1, Lltesin 64)
i= (18)
“The graph of for 0 =< 4, given in FIGURE 445, was obtained with the help of aCAS. =
nswers to selected odd-numbered problems begin on page ANS-9.
rms In Problems 9-14, use the Laplace transform to solve the given
initial-value problem, Use the table of Laplace transforms in,
To evaluate the given Appendix IIT as needed,
Hee} 8 y ty=rsing, =O
Ef rsinb 31} 10. y’-y=e'sins, yO) =0
E{F cos 1) M1, y+ 9y = c0834, yO) = 2, »’(O)
Ste" cos 34) @ yty=sintg yO)=1, y=
44 Additional Operational Properties | 245 1 y+ 16y =f, 0) =
b {co
KO = Yo.
1 y+ y=, 0) = 1, yO) =O, where
1 Os1< aR
In Problems 15 and 16, use @ graphing utility to graph the
indicated solution,
15. y(0) of Problem 13 for 0
16. (0) of Problem 14 for 0
In some instances the Laplace transform can be used to solve
linear differential equations with variable monomial
coefficients. In Problems 17 and 18, use Theorem 4.4.1 to
reduce the given differential equation toa linear first-order DE
in the transformed function ¥(s) = ££(y()}. Solve the first-order
DE for Y(s) and then find ya) = £-"(¥())
1, "y= 20, (0) =0
18, 2y" +1" —2y= 10, y(O)=y'@ =0
<2
13m
442 Transforms of Integrals
In Problems 19-22, proceed as in Example 3 and find the
convolution f+ g ofthe given functions. After integrating find
the Laplace transform of f+
12 f= 41,80 = 37 2 fe)
A. f= ee =e fi)
In Problems 23-34, proceed as in Example 4 and find the Laplace
transform of f+ g using Theorem 4.4.2, Do not evaluate the
convolution integral before transforming.
2B. Lee} 2 L(P ete!)
B. Ele ee cost} 25. Ele sin)
niffeat wef {wsre}
a offercsra} om a{frsnra}
na{ferw} 2 [secon o}
se afi[inra} }
In Problems 35-38, use (8) to evaluate the given inverse transform.
Be tz - a} * eae a
seta} = etal
438, ‘Te table in Appendix II does not contain an entry for
ys { Bis 3}
ere
246 | CHAPTER 4 The Laplace Transform
9) =
0821, ) = &
(a) Use (4) along withthe result in (5) to evaluate this inverse
transform, Use a CAS as an aid in evaluating the convo-
lution integral
(b) Reexamine your answer to part (a). Could you have
obtained the result in a different manner?
4. Use the Laplace transform and the result of Problem 39 0
solve the initial-value problem
Yity=sinetrsins, yO) =0, yO =0.
Use a graphing utility to graph the solution
In Problems 41-50, use the Laplace transform to solve the given
integral equation or integrodifferential equation
a. po+ [e~onmar=
a f= e+ | afte — nde
a. f(t) + 2| fox)coste — 2)dr = de" + sine
© 0+ [node =
spo = anes [ery =na
of — 141 8f = oye
fo 15 L 0 9% nar
1 7 = [ee OE
fseoar, 0) = 0
4. y= 1 = sine —
ay
+ 60 + 9[ yar = 1, 0) =
GO +9| sede = 1, yO) =o
In Problems 51 and 52, solve equation (10) subject to i(0)
with Z, R, C, and £(@) as given. Use a graphing utility to graph
the solution for 0=1= 3,
BL =0.1b,R=3.0,C= 0058,
(9 = 100 = 1) = UE = 2)]
82 L=0.005h, R= 10,C= 0.024,
F( = 100[1- @ = NAUE= 1)
53. The Laplace transform ‘£{e~")
solve the initial-value problem
exists, but without finding it
y+ 9y = 37%, 0) = 0, ¥"0)
54, Solve the integral equation
fo~ e+e eorar 4
In Problems 55-60, use Theorem 44.3 to find the Laplace
3 Transform of a Periodic Function
transform of the given periodic function.
Tt
]
ay tay Sa ey
Meander funtion
FIGURE 445 Graph for Problem 55
fo
FIGURE 44.7 Graph for Problem 36
fy
bab 3
Saoth fonction
FIGURE 448 Graph for Problem $7
so)
Triangular wave
FIGURE 44 Graph for Problem 58
VY
© Ww an ‘
FIGURE 4.410 Graph for Problem $9
fo
oe ae ar
FIGURE 44.11 Graph for Problem 60
In Problems 61 and 62, solve equation (16) subject to (0) = 0
with £( as given. Use a graphing utility to graph the solution
for 0 = = 4 in the case when L = 1 and R= 1
61. (0 is the meander function in Problem 55 with amplitude 1
anda = 1
62. (isthe sawtooth function in Problem 57 with amplitude 1
and b = 1
In Problems 63 and 64, solve the model for a driven spring/mass
system with damping
a “1
+ be =f), 0) = 0, x0) = 0,
‘where the driving function fis as specified. Use a graphing
utility to graph x( for the indicated values ofr.
B . Fis the meander function in Problem $5
with amplitude 10, and a = ,0 = ¢= 2m,
64. m= 1,8 =2,k= 1, fis the square wave in Problem 56 with
amplitude 5, anda = 7,01 4x.
Discussion Problems
1S. Show how to use the Laplace wansfonn to find the numerical
Value of the improper integral | te~™sin4rd
66. In Problem 53 we were able to solve an initial-value problem
without knowing the Laplace transform £(e™"}.Inthis prob-
Jem you are asked to find the actual transformed function
¥() = £le™"} by solving another initial-value problem.
(a) Ify = e°*, then show that y isa solution of the inital-
value problem
»
+ (0) = 1
(b) Find ¥(s) = £{e7"} by using the Laplace transform to
solve the problem in part (a) [Hin: First find ¥(0) by
rereading the material on the error function in Section 2.3
‘Then in the solution ofthe resulting linear first-order DE.
in ¥(3) integrate on the interval [0,5]. also helps to use a
dummy variable of integration. ]
67. Discuss how Theorem 4.4.1 can be used to find
(0
68, Bessel’s differential equation of order n
yl ty ty=0.
‘We shall seein Section 5.3 thata solution of the nital-value prob-
Jems” +y' +19 = 0,310) = 1,y'O) = Disy = (0, called the
‘Bessel function ofthe first kindof order» = 0, Use the procedure
‘outlined in the instructions to Problems 17 and 18 to show that
[Hint: You may need to use Problem 52in Exercises 4.2. Also,
iis known that Jg(0) = 1]
44 Additional Operational Properties | 247 's differential equation
p+ (= 2y’ tny=0
is known to possess polynomial solutions when m is a
nonnegative integer. These solutions are naturally
called Laguerre polynomials and are denoted by L,()
Find y = L,(0, forn = 0, 1,2, 3, 4 if itis known that
1,0) = 1
(b) Show that
{££ pe} on
fh v6,
where H(s) = SE(y) andy = L,(isa polynomial solution
ofthe DE in par (a). Conclude that
artes n= 001.2,
LO =
Computer Lab Assignments
71. In this problem you are led through the commands in
Mathematica that enable you to obtain the symbolic Laplace
transform ofa differential equation and the solution ofthe
initial-value problem by finding the inverse transform. In
Mathematica te Laplace transform of afunction (is obtained
sing LaplaceTransform [yt},s} laline wo ofthe syntax, we
replace Laplace Transform [y[t]t,$] by the symbol ¥. (Ifyou
do not have Mathematica, then adapt the given procedure by
{finding the corresponding syntaxfor the CAS youhave on hand.)
Consider the initial-value problem
yt + Gy" + 9y =esint, (0) =2, yO) =—1
Precisely reproduce and then, in turn, execute each line in the
given sequence of commands, Either copy the output by hand
fr print out the results,
diffequat = y"It] + 6y'It] + 9ylt] == tSin{t]
transformdeq = LaplaceTransform [diffequat, , s/.
{yl0] — >2,y'10] - >—1,
LaplaceTransform [yt] t, ] — > Y)
soln = Solve(transformdeg, ¥] // Flatten
Y=Yi.soln
InverseLaplaceTransform!Y, s,t]
TI, Appropriately modify the procedure of Problem 70 to find a
solution of
y+ 39 =4y=0, 1) = 0. 'O =O, YO=1
72 The charge (9 ona capacitorin an LC-seies circuits givenby
a
Sat a= 14M m) + 60 — 30), 40) = 0,4") =
Appropriately modify the procedure of Problem 70 to find
4). Graph your solution.
() Behavior of 3, 82-90
FIGURE 45.1. Unit impulse
248
45 The Dirac Delta Function
INTRODUCTION Just before the Remarks on page 223, we indicated that as an immediate
consequence of Theorem 4.2.3, Fs) = | cannot be the Laplace transform ofa function that is
Piecewise continuous on [0 0) and of exponential order, In the discussion that follows we are
{going to introduce a function that is very different from the kinds that you have studied in previ-
‘ous courses. We shall see that there does indeed exist a function, or more precisely a generalized
function, whose Laplace transform is Fs) = 1
I. Unit Impulse Mechanical systems are often acted on by an external force (or emf in an
cleetical eicuit) of large magnitude that aets only for a very short period of time, For example,
vibrating airplane wing could be struck by Iightning, a mass ona spring could be given a sharp
blow by a ball peen hammer, a ball (baseball, golf ball tennis ball) could be sent soaring when
struck violently by some kind of club (baseball bat, golf club, tennis racket), The graph of the
piecewise-defined function
0St< 4
Br w=
0,
t 0, f> 0, shown in FIGURE 45a), could serve as a model for such a fore. Fora small value of a,
{8,0 ~ 4) is essentially constant function of large magnitude that is “on” for just a very short period of
time, around fy, The behavior of 3,(¢ ~ f)) as a> is illustrated in Figure 4.5.1(6). The function
£5,(¢— t) iscalled « unit impulse sine it possesses the integration property J58,(t — 4) dt = 1
| CHAPTER 4 The Laplace Transform L The Dirac Delta Function In practice itis convenient to work with another type of unit
impulse, a “Function” that approximates 8,(¢— f) and is defined by the limit
Ble — 9) = lim 3,0 ~ 1. 2
‘The later expression, which is nota function at all, can be characterized by the two properties
oo T=
o a=) {* +e
‘The unit impulse 8(1 ~ f) is called the Dirac delta function.
tis possible o obtain the Laplace transform of the Dirac delta function by the formal assump-
tion that ££(8(1— 69)} = lim, 2(8,(¢~ 19)
nee
For fy > 0, £{8(t - ) = eo (3
See
PROOF: To begin, we can write ,(t ~ 4) in terms of the unit step function by virtue of (11) and
(12) of Section 4.3
1
BAL 19) = 52 [UE (by = a) — UE = Cop + 2]
By linearity and (14) of Section 4.3, the Laplace transform of this last expression is
1
t= oh = 3S
Since (4) has the indeterminate form 0/0 as a> 0, we apply L’Hpital’s rue:
(60 = wh = tim 1840~ g) = tin ( TEE) =
‘Now when fy = 0, it seems plausible to conclude from (3) that
£(6@) = 1.
‘The last result emphasizes the fact that 8() is not the usual type of function that we have been
‘considering, since we expect from Theorem 4.2.3 that £((#)} 0 as s ->00,
‘Two Initial-Value Probl
Solve y' + y = 46(0 — 2m) subject to
@ 9O=1, YO=0 — ) Y=, Y@=0.
‘The two initial-value problems could serve as models for describing the motion of a mass on
1 spring moving in a medium in which damping is negligible. At ¢ = 2m the mass is given a
sharp blow. In part (a) the mass is released from rest | unit below the equilibrium position. In
part (b) the mass is at rest in the equilibrium position,
SOLUTION (a) From (3) the Laplace transform of the differential equation is
2M) — 54 Ho) = 4e™ oF Y)= AT A se
‘Using the inverse form of the second translation theorem, (15) of Section 4.3, we find
MO = cos + 4 sin(t — 2m) UCr — 2m.
45 The Dirac Delta Function | 249 FIGURE 452 In Example 1(2), moving
mass is struck at = 2
FIGURE 45.3 In Example 1), mass is at
rest until struck at = 2
Since sin(e — 2m) = sin the foregoing solution can be written as
cost O51<2n
0 oe aune nae 8
In FIGURE 45.2 we see from the graph of (5) that the mass is exhibiting simple harmonie motion
until it is struck at ¢ = 277, The influence of the unit impulse is to increase the amplitude of
vibration to VIT for #> 2,
(b) In this case the transform of the equation is simply
det"
¥) =
and so MK) = dine 2m) U2)
- (0
~ La sins,
The graph of (6) in FIGURE 4533 shows, as we would expect from the initial conditions, that
the mass exhibits no motion until it is struck at = 27.
6)
EEE
REMARKS
(1 8(¢ ~ &) were a function in the usual sense, then property (i) on page 249 would imply
SF8C — t) dt = Orather than fi°5{¢ ~ f) dt = 1. Since the Dirac delta function did not “behave”
like an ordinary function, even though its users produced correct results, itwas met initially with
great scorn by mathematicians. However, in the 1940s Dirac’s controversial function was put
‘on a rigorous footing by the French mathematician Laurent Schwartz in his book Théorie des
distributions, and this, in (umn, led to an entirely new branch of mathematies known as the
theory of distributions or generalized functions. In this theory, (2) isnot an accepted definition
‘of (¢~ &), nor does one speak of a function whose values are either o0 or 0. Although we shall
‘not pursue this topic any further, suffice it to say that the Dirac delta function is best character-
ized by its effect on other functions. Iffis a continuous function, then
{ SOB — tp) dt = fa) )
ccan be taken as the definition of 8(¢ — %.). This result is known as the sifting property since
{8(¢— fq) has the effect of sifting the value f(t.) out of the set of values of fon [0, 00). Note that
property (i) (with f(7) = 1) and (3) (with f(t) = e~*) are consistent with (7).
(ii) In the Remarks in Section 4.2 we indicated that the transfer function of a general linear
nh-order differential equation with constant coefficients is W(s) = 1/P(s), where P(s) =
as" + a_,8"~! +" + ay, The transfer function is the Laplace transform of function w(t),
called the weight function of a linear system. But w(#) can be characterized in terms of the
discussion at hand. For simplicity let us consider a second-order linear system in which the
input is a unit impulse at
ayy" + ayy’ + amy = 8), yO)=0, yO) =0.
Applying the Laplace transform and using ‘£(8(®)} = 1 shows that the transform of the
response yin this case is the transfer function
¥%@) = = = 4 = We andso yoo
- -
aetasta, PO
= wo.
ait ee
From this we can seo, n general, that the weight fanetion y = w(t) of an nth-order Hines
systemis the zero-sate response ofthe system oui imps For this eason w() scaled
2 wel the impulse response of the system
——Ss—
250 | CHAPTER4 The Laplace Transform HGR BIS _Answers to selected odd-numbered problems begin on page ANS-10.
In Problems 1-12, use the Laplace transform to solve the given
differential equation subject to the indicated initial conditions.
Ly 3y=8U- 2, =O
2 yty=8-D, yO=2
3 y+ y= 602m), 0) =0,y'0)=1
4. y' + 16y = 802m), (0) = 0,» = 0
5. y+ y= (r= wI2) + 8-302), XO) = 0,9") =0
6 y'+ y= 602m) + 50¢—4m}, (0) = 1,y'O =O
Ty #2) = 84=D, = O,y'O = 1
8 y'—2y'= 1480-2, YO =0.y'O)
9. y' +4)" + 5y= 80 -2n), 10) = 0.9") =0
10. y" + 2y + y= —E-D, yO) = 0.9") =0
My" +4y' + 1By = 50 m) +6¢- 3m), OV =1y'O=0
12 yy’ +6y= +32) + 50-4), WH =0.¥O=0
In Problems 13 and 14, use the Laplace transform to solve the
Graph your solution on the interval
given initial-value proble
[0, 87],
Bye
Dae - kx), (0) = 0,y'@ = 1
yt ye Fae — 2m, yO = 090 =1
In Problems 15 and 16, a uniform beam of length L carries @
concentrated load ws at x = $2. Solve the differential equation
“
a1 = mae
z .OSe