0% found this document useful (0 votes)
8 viewsdifferential equation
differential equation
Uploaded by
Mostakim HossainCopyright
© © 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
0% found this document useful (0 votes)
8 viewsdifferential equation
differential equation
Uploaded by
Mostakim HossainCopyright
© © 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
/ 165
Differential Equations I
MATB44H3F
Version September 15, 2011-1949 Contents
1 Introduction
11 Preliminaries
12 Sample Application of Ditfe
ial Equations
2. First Order Ordinary Differential Equations
21 Separable Equations
2.2. Exact Differential Equations
23 Integrating Factors
24 Linear First Order Equations
25. Substitutions
25.1 Bernoulli Equation
2.8.2. Homogeneous Equations
Substitution to Reduce Second Order Equations to First
Order
3. Applications and Examples of First Order ope’s
3.1 Orthogonal Trajectories. .
3.2 Exponential Growth and Decay
33 Populstion Growth
34 Predator-Prey Models
35 Newton's Law of Cooling
3.6 Water Tasks és
3.7 Motion of Objects Falling Under Gravity with Air Resistance
3.8 Escape Velocity
3.9 Planetary Motion
3.10 Particle Moving on a Curve
25
25
2
28
29
20
aL CONTENTS
Linear Differential Equations 45,
4.1 Homogeneous Linear Equations
4.1.1 Linear Diffocential Equations with Constant Coofficents
4.2 Nonhomogeneous Linear Equations .
Second Order Linear Equations 87
5.1 Reduction of Order... 6. « ssa veces: SBP
5.2 Undetermined Coefficients . cect eee eter e ees 60
5.2.1 Shorteuts for Undetermined! Coefficients 64
5.3 Variation of Paramoters . : . 66
Applications of Second Order Differential Equations 1
6.1 Motion of Object Hanging from a Spring 7
6.2 Blectrical Cireuits 6. eee eee eee ees TB
Higher Order Linear Differential Equations: 70
7.1 Undetermined Coefficients 9
7.2 Variation of Parameters... 0.6 eevee eee 80
7.3 Substitutions: Euler's Equation : sz
Power Series Solutions to Linear Differential Equations 85
8.1 Introduction... — g 89
8.2 Background Knowledge Concerning Power Sories 88
83° Analytic Equations... 6. 89
84 Power Series Solutions: Levels of Success... ee ee OL
8.5 Level I: Finding a finite number of coefficients a1
6 Lovel 2: Finding the recursion relation o
8.7 Solutions Near a Singular Point . 9 meee OY
8.8 Functions Defined via Differential Equations au
88.1 Chebyshev Equation... 6.062220 eee eee) dn
882 Legendre Equation us
883° Airy Equation 22... cee MS
884° Laguecre's Equation»... . eee eee UB
88.5 Bessel Equation 116
Linear Systems 121
9.1 Preliminaries. . “mys z seve wo RL
9.2, Computing e™ 128
9.3 The 2x 2 Case in Detail = : ewe
9.4 The Non-Horogencous Case... serresraranee sstcace IOS CONTENTS ¥
9.5 Phase Portraits 6. e188
9.5.1 Real Distinet Eigenvalues 37
9.5.2 Complex Eigenvalues 139
9.5.3 Repeated Real Roots Ml
10 Existonce and Uniqueness Theorems 145
10 Picard’s Method oo... ee eee eee nas aon 1B
10.2 Existence and Uniqueness Theoren for First Order ODES. . . . 150
10.8 Existence and Uniqueness Theorem for Linear First Order ODE’s 155
104 Existonce and Uniqueness Theorem for Linear Systems 156
11 Numerical Approximations 163
111 Euler's Method 2 . 63
HAA Error Bounds 0... eee es 165
11.2 Improved Euler's Method... sac ao waa 1
11.3 Runge-Kutta Methods 167 CONTENTS Chapter 1
Introduction
1.1 Preliminaries
Dofinition (Differential equation)
A differential equation (pe) is an equation involving a function and its deriva
tives,
Differential equations are called partial differential equations (PDB) or or
dinary differential equations (ODE) according to whether or not they contain
partial derivatives. ‘The order of a differential equation is the highoot order
derivative occurring. A solution (or particelar solution) of a differential equa
tion of order n consists of a function defined and n times differentiate on a
domain D having the property that the functional equation obtained hy substi-
tuting the fumetion and its n derivatives into tho differential equation holds for
every poiat in D.
Example 1.1. An example of a differential equation of order 4, 2, and 1 is
given respectively by
.
2)
sin(a) + 605%(2), CHAPTER 1. INTRODUCTION
Example 1.2. The function y = sin(2) is a solution of
(2) + EE y= 2sins) +c0s4(a)
‘on domain R; the function 2 = 6 cos(y) is a solution of
Pe ae
ost” OF
on domain R?; the function y = 2V is a solution of
‘on domain (0,00).
Although it is possible for a Dx to have a unique solution, eg. 9
solution to (y/)* +y? = 0, or no solution at all, eg. (y/)? +9? =
itely many solutions.
solution, most DE’s have
Example 1.3. The function y= Ve+C on domain (~C/4, 00) isa solution
of yy! = 2 for any constant C. ‘
Note that diferent solutions can have different domains. The set ofall
solutions to seal fs gen solution,
1.2 Sample Application of Differential Equations
A typical application of differential equations proceeds along these lines:
Real World Situation
1
Mathematical Model
1
Solution of Mathematical Model
1
Intorpretation of 1.2. SAMPLE APPLICATION OF DIFFERENTIAL EQUATIONS 3
Sometimes in attempting to solve a DE, we might, perform an irreversible
step. This might introduce extra solutions. If we ean get a short list whieh
contains all solutions, wo can then test out each one and throw out the invalid
ones. The ultimate test is this: does it satisfy the equation?
Here isa sample application of differential equations,
Example 1.4. The half-life of radium is 1600 years, L., it takes L000 years for
half of any quantity to decay. If a sample initially contains 50g, how long, will
it be until it contains 45.0? .
Solvtion. Let 2(t) be the amonnt of radinm present at time tin years. The rate
at which the sample decays is proportional to the size of the sample at that
time. Therefore we know that d/di = kz. ‘This differential equation is our
mathematical model. Using techniques we will study in this course (soe §3.2,
Chapter 3), we will discover that the general selution of this equation is given
constant A. We are told that = 50 when
by the equation «= Ae, for som
1 = 0 and so substituting gives A = 80. Thus 2 = 50e!. Solving for t gives
1 = In(x/50) /k. With (1600) = 25, we have 25 = 501%, ‘Thorofore,
'
sone=s(!)
siving us k = —In(2) /1600. When » = 45, we have
142/90) __ 945/90) ggg, UBIO, In(u0/8)
E ~~ —In(@)/1600 Tn) Tn)
0.105,
= 1600.0
1600 x 0.152 ~ 249.9,
‘Therefore, it will be approximately 243.2 years until the sample contains 45.
of radium. °
Additional conditions requized of the solution (2(0) = 50 in the above ex-
ample) are called boundary conditions and a diferential equation together with
boundary conditions is called a boundary-value problem (BVP). Boundary con-
ditions come in many forms. For example, y(6) = y(22): y!(7) = 3y(0); y(9) = 5
are all examples of boundary conditions. Boundary-value problems, like the one
in the example, where the boundary condition consists of specifying the value
of the solution at some point are also called initial-value probloms (IVP)
Example 1.5. Au aualogy from algebra is the equation
ya ver2 (44) 4 CHAPTER 1. INTRODUCTION
‘To solve for 1, we proceed as
(reversible step)
(y— (y- 4) =0.
‘Thus, the set y € {1,4} contains all the solutions. We quickly see that y = 4
satisfies Eaquation (1.1) because
daVit2—ea
while y= I does not because
1=V142—51=3
So wo accept y= 4 and reject y Chapter 2
First Order Ordinary
Differential Equations
‘The complexity of solving DE’s increases with the order. We begin with first
order DE’s.
2.1 Separable Equations
A first order ope has the form F(2,y, yf) = 0. In theory, at lest, the methods
of algebra can be used to write it in the form* y= G(e,y). If G(e,y) con
be factored to give G(x.4) = M(x) N(y)the
To solve the separable equation y’ = M(x) N(y), we rewrite it in the form
Flu)y! = 912). Intograting both
[sorta = f s6e)ac,
[rorar= [sar ac.
Example 2.1. Solve 2ry +60 + (1? — 4) =0. *
quation Is called! separable
les gives
We ase the notation dy/de = Ge) aad dy = Gls, y)de interchangeably 2.2, EXACT DIFFERENTIAL EQUATIONS 7
An alternate method to solving the problem is.
dy = ~sin(x) de.
[vie ff say
cos(1) ~ cos(0),
Fela) -1
© casts) 3,
Teese)
giving us the same result as with the first method °
Example 2.3. Solve w'y’ +4 +22 +1
Solution, We have
where C is an arbitrary constant, This is an implicit solution which we cannot
easily solve explicitly for y in vermns of x. °
2.2 Exact Differential Equations
Using algebra, any frst order equation can be written in the form F(,y) de +
(2,4) dy = for some functions F(x, y), Gla, y).
Definition
An expression of the form F(z, y) da +G(x,1) dy is called a (first onder) differ
ential form, A differentical form F (x,y) dr + G(x,y) dy is called exact if there
exists a function g(z,y) such that dy = Far + Gedy.
Ifw = Fde-+G dy is an exact differential form, then w = 0 is called an exact
differential equation. Its solution is g = C, whore w = dg.
Recall the following useful theorem from MATB42: SCHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
‘Theorem 24
IF and G are functions that are continuously diferentiable throughout a
simply commected region, thon F dz +C dy is exact if and only if AG/r
OF oy
Proof, Proof is given in MATBA2. o
Example 2.5. Consider (82x? +22) dx + (2xy + 92) dy = 0. Lat
w= (B07? +27) de (2a%y + 4?) dy
ane
‘Then note that
ar
=6cty= OF
ee Oy
Or
By Turonen 2.4, w = dy for some g. To find g, we know that
(22a)
(1b)
22) 2.2, EXACT DIFFERENTIAL EQUATIONS 9
for some arbitrary constant C. ‘Therefore,
‘quation (2.2) becomes
gay +
Note that according to our differential equation, we have
sya ®4.¥ 40) <0 which tmption 2%)? 42
a(evs P4240) =o oties implies ot 4 2 +
#
P+e=c'
for some arbitrary constant
constant, the solution is
"Letting D = C’ ~ C, which is still an arbitrary
mr+>
ai
s
Example 2.6. Solve (32 + 22y") dz + (22%y) dy = 0, where y(2)
Solution. Wo have
Statyt +o
i (322 + ey?) de
for some arbitrary constant C. Since C is arbitrary, we equivalently have 23 +
2%y? = Q. With the initial condition in mind, we have
8449-0 0-4.
Therefore, 23 + 24y? = 4 and it follows that
svat
——a
But with the restriction that y(2) = 3, the only solution is
waa
wes
on the domain (~YH, YH) \ {o}. °
Let w = Fede +Gdy. Let y = s(a) Ve the solution of the De w = 0, Le,
F + Gs!(x) =0. Let no = (00) ancl let + be the piece of the granhh of y = s(2)
from (x0,p) to (2,y). Pisgure 2.1 shows this ides. Since y = s(x) is a solution
tow = 0, we must have w = 0 along 7. Therefore, fw = 0. This ean be seen 10CHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS.
y
Figure 2.1: The graph of y = (2) with + connecting (20, uo) to (2,9).
by parameterizing by 9(r) = (2, (2)), thereby giving us
[or [[Fucseeqae= J oto.
‘This much holds for any 2.
Now suppose that w is exact, Then the inteural is independent of the path,
‘Therefore
on fon [rasa cays f récsoe
-[ Gen nay+ f Flay)de.
‘We can now sclve Example 26 with this new method.
Solution (Alternate solution to Example 2.
|. We simply have
‘ z
om [2-2udu J (a0 +20?) ae
4g? —4 (C3) +29 + 24? — 29 28?
dy? — 36+ 2° +27y? — 8 — ay?
finally giving ws 23 + 2%y = 44, which agrees with our previons answer.
Remark. Separable equations are actually a special ease of exact eqquations, that
a
ay 9)
Hea = a2) = a(x) dr + sy) eu =0=> 240) = 2.9. INTEGRATING FACTORS "
So the equation is exact. °
2.3 Integrating Factors
Consider the equation w = 0. Bven if w is not exact, there may be a fanetion
T(z, y) such thot Iu is exact. $0 4 = 0 can be solved by multiplying both sides
by I. The function F is called an integrating factor for the equation w = 0.
Example 2.7. Solve y/2?+1+u/a
Solution. We have
la: (2)
So the equation is not exact. Multiplying by 2? gives us
Wo sve that
(y+2?) da + rdy=0,
a(y+ 2) =o
id
w+e-0
for some arbitrary constant C. Solving for y finally gives us
mes °
‘Thore is, in general, no algorithm for finding integrating factors. But the
following may suggest where to look. Tt is important to be able to recognize
common exact forms:
ady + ydx = d(xy),
zr=v4e _ (2)
de ala? 4-42)
cere [ee st ):
aor = a(en(2)),
—" (ay de + bx dy) = d(a"y') 12CHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
Example 2.8. Solve (c*1? + 9) dr + (22%y ~ x) dy =0. *
Solution. Expanding, we have
ay? de + 2etydy + yde—edy
Here, a= 1 and b= 2, Ths, we wish to tse
aay?) = y? de + 2ry dy.
‘This suggests dividing the original equation by 2? which gives
de rdy _
war 2eydy + MOS
0.
‘Therefore,
ay t
Cc, x40,
where C is an arbitrary constant. Additionally, y = 0 on the domain R is a
solution to the original equation °
Example 2.9. Solve yd — 2 dy — (x? + y2) de = 0 *
Solution. We have
y=ztau(D—2),
Where C is an arbitrary coustant and the domain is
D-eF Qnty 5, e¥@ntyt
for any integer n. Also, since the derivation of the solution is based on the
assumption that 2 £0, it is uncloar whether or not 0 should be in the domain,
ive, does y = xtan(D — 2) satisfy the equation when 2 = 0? We have y — 2y/— 2.9. INTEGRATING FACTORS 13
(22 +4) =0. x= 0 and y=atan(D~2), then y = 0 and the equation is
satisfied. ‘Thus, 0 is in the domain. °
Proposition 2.10
Let w= dg. Then for any funetion P : R + R, P(g) is exact.
Proof. Let Q = J P(t) dy, Then d(Q(g)) = P(g) dg = Pgh. a
To make use of Proposition 2.10, we can gromp together some terms of w
to got an expression you recognize as having an integrating factor and maukiply
the eqjtation by that. ‘The equation will now look like dg + = 0. If we can
find an. integrating factor for h, it will wot necessarily help, ince multiplying by
‘it might mess up the part that is already exaet, But if we can find one of the
form P(g), thon it will work.
Example 2.11. Solve (0 ~ ys") dy + yd
Solution. Expanding, we have
yde+ dy —yx? dy =0.
eee:
ae
‘Therefore, we ean multiply teh equation by any function of xy without disturb-
ing the exactness of its first two terms. Making the last term into a function of
y alone will make it exact, So we multiply by (ry) ?, giving us
udetady 1
ta(lv) =
aP (lu) =C.
where C’is an arbitrary constant, Note that y = 0 on the domain R is also a
solution, °
Given
Mdx+Ndy=0, (o)
‘we want to find f such that 1M dr + IN dy is exact. If s0, then
a a
Zun-20
oe
ENTS TAIN, AACHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS.
If we can find any particular solution {(c, 9) of the PDE
LN +IN,
M+IM,, (8)
then we can use it as an integrating factor to find the general solution of (+)
Unfortunately, (++) is ueually even harder to solve than (2), but as we shall sco,
there are times when itis easier.
Example 2.12. We could look for au J having only 2's and uo y's? For exam
ple, consider fy = 0. Then
LN -+1N,= ie 2s — MMe
This works if (My ~ Nz) /N happens to be a function of ¢ alone. Then
reef ttgteae
‘Similarly, we ean also reverse the role of and y. If (Ne —My)/M happens to
bbe a fanetion of y alone, then
of Satey
works *
2.4 Linear First Order Equations
AA first order linear equation (n1 = 1) looks like
vy +Pay
(2),
An integrating factor can always be found by the following method. Consider
dy + Ple)ydr = Qed,
(Pxy Qa) de + dy, =0.
Cer kee S
Mea) Now)
We use the DE for the integrating factor F(a, y). ‘The equation FA d+ IN dy
is exact if
IN + IN, = 1M +IM,. 24, LIN
EAR FIRST ORDER EQUATIONS 15
Im our ease,
1. +051, (Ply —Qla)) +1P(e), (©)
Wo noed only one sclution, so wo lock for one of the form I(2), io., with J
Then (*) becomes
at ‘
Fa IPC)
his is separable. So
ar
4 = pear,
man = freerse,
=f, &>0
ef Pe) te
Wo conclude that ef P14 is an intograting factor for + P(x) = Q(2)
Example 2.13. Solve y/ ~ (I/r)y =, where x > 0. *
Solution. Here P(x!
=1/x. Then
Teel Pelt gS 84e gine | ED
fel ~ >
where 2 > 0. Our differential equation is
ady—ydr
re re
Multiplying by the integrating factor 1/s gives us
zdy~yde
z
wader.
Then
ou the domain (0,50), where Cis an arbitrary constant (2 > (is given). 1GCHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
In general, given 9! + P(x)y = Q(a), multiply by e! ” + to obtain
el Pla)
tel POD4* Piady = Qlael Pe,
areas
Therefor,
wel 4 = fear + 6,
yee IPM fatal Porta a co FPO,
whore Cis an arbitrary constant.
Example 2.14. Solve xy’ +2y = 42" *
Solution. What should P(x) be? ‘To find it, we put the equation in standard
form, giving us
ig ie,
u + I =e.
‘Therefore, P(x) = 2/-r. Immediately, we have
T= olteizide — giz”)
Multiplying the equation by 2? gives us
where C is an arbitrary constant and x # 0. °
Example 2.15. Solve e°? dy + dr + 2edy =
Solution. This equation is linosr with 2 as a function of y. So what we have is 2.5, SUBSTITUTIONS 7
where I =e! *4¥ =e, Therefore,
a
ert 5 net = 8,
at
-e +6,
where C is an arbitrary constant, We could solve explicitly for x, but itis unessy.
‘The domain is not easy to determine °
2.5 Substitutions
In many cases, equations can be put into one of the standard forms discussed
above (separable, linear, ete.) by a substitution,
Example 2.16. Solve yf’ ~ 2y' =, *
Solution. This is a frst order linear equation for y/. Let u = y'. Then the
equation becomes
wv
Tho integration factor is thon I = o~ S24! m o-2!, Thus,
wee? — 2ue-® = 5o-%
where C is an arbitrary constant. But u=
Be
grt GOP prt Ce tC
on the domain B, whore C, and C; are arbitrary constants, °
‘We now look at standard substitutions.
2.5.1 Bernoulli Equation
‘The Bernoulli equation is given by
+ Pla)y = Q2)y".
le dy
Erte ASCHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
ving us
4
om
&
1
+ Play
a
Tondr
& + 0-mP@==01-mMacr.
Oe),
+ P@)z=Q2),
‘which is linear in =
Example 2.17. Solve y! + 2
Solution. Here, we have a= 3. Let z=y-*. ify #0, then
We can readily see that J =
‘The domain is,
RB, e>
Ij > V=mHO), Cs-.
An additional solution is y =O on B. ° 2.5, SUBSTITUTIONS 19
2.5.2 Homogeneous Equations
Definition (Homogeneous function of degree m)
A finetion F(.r,y) is called homogeneous of degree m if F(x, 4y) = X"F(2,y)
For a polynomial, homogencous says that all of the terms have the same dogree.
Example 2.18. The following are homogeneous functions of various degrves:
48+ 524? — homogeneous of degree 6,
Sa® 452% not homogeneous,
2VF EF homogencans of degree 2,
sn(4) —omopenans of dee
Fey Remozeows of degree —1 *
IEF is homogeneous of degree n and C is homogeneous of degree k, then
F/G is homogencous of degree n~ k.
Proposition 2.19
If F is homogencous of degree 0, thea F is a function of y/s
Proof. We have F(Ax,Ay) = F(2,y) for all A. Let = A/z. ‘Then Flay) =
Fllu/2). a
Example 2.20. Hero are some examples of writing a homogeneous function of
degree 0 as a function of y/a.
w/a) + (u/2)
vtaty
aye (w/a +
Consider M(z,y) de + N(2,y) dy = 0. Suppose M and 1 are both homoge-
neous and of the same degre. Then 20CHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
‘This suggests that v= y/z (or eqmivalently, y= vr) might hep. In fact, write
‘Then
= Riv).
‘Therefore,
dv
oe = Reo) —
dy de
Rw-0 a
which is separable, We conclude that if M and N are homogeneous of the same
degree, setting y = ux will give « separable equation in v and .
Example 2.21.
Solve ry! dy = (x? +y3) de «
Solution. Let y= ex. Then dy = vdz + 2dv, and our equation becomes
avta2 (de + adv) a4) dr,
8 det 240? dy = 29 de +02? do
‘Therefore, x = 0 or v? dv = dr/z. So we have
© = tn({z/) +0 = tnx) + in(\Al) = In({Ae)) = In( Ar).
3 “
where the sign of A is the opposite of the sign of 2. ‘Therefore, the general
solution is y = x($ln(Az))'/*, where A is a nonzero constant, Every A > 0
‘yields a solution on the domain (0,00); every A < O yields a solution on (—o, 0).
Jn addition, thore is the solution y = 0 on the domain R. 6
2.5.3 Substitution to Reduce Second Order Equations to
First Order
A second order DE hus the form
Plu" y'sy.2) 2.5, SUBSTITUTIONS 21
If it is independent of 1, namely, F(y",y/',22) = 0, then it is really just a first
order equation for y/ as we saw in earlier examples,
Consider now the case where it is independent of x, namely, F(y/",y’.y) = 0.
ubstitnte » = dy/dr for x, ie, eliminate 2 between the equations
and » = dy/de. Then
‘Therefore,
v)-0-
‘This is a first order equation in and y.
Example 2.22. Solve y/' = 4(y/)"”* y, *
Solution, Let v= dy
dx. Then
fy _ de
way
and our equation becomes
ao
de 408/y,
dy "
dy
T= syd, v>0,
a7 ta
2B = WP +O,
vi=y" +Cx 22CHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
whore Cy is an arbitrary constant and C2 =C1/2. But v = dy/dr., so we have
dy 2
ae tay
ten
Pray
fos
vray
sar (ww (Fe) + FR) +e Goo,
=) dt =0, °
marae wi ten Go.
‘Next consider second ocder linear equations. That is,
Play" +Q@)u' + Ray = 0.
We can eliminate y by letting y =". Then y =e"! and yf" =e" (0)? +e"
‘The equation then becomes
Phe) (ee)? Hee") + Qlahe + Rlele —0
which is a first onder equation in of.
Example 2.28. Solve 2%y/" + (x —22) yey =0. *
Solution. Let y =e, Then the equation bocomes
27° (vw)? + 2teu" + (w— 2") eu! — ee”
White «
Then
we +2829 + (1-2) 2
Now we are cn our own—there is no standard technique for this case. Suppese
wo tey w= ay. Then + = w/z and
‘Then it follows that
au! tu? 20
‘This is a bit simpler, but itis still uonstandard. We ean see that letting u = ae 2.5, SUBSTITUTIONS 23
will give us some cancellation. Thus, u’ = s'e*-+ se* and our equation becomes
ae base + Pe* — ae =F,
asltetet ae,
Working our way back through the subsitutions we find that = zre~* so our
solution becomes
fue
Using algebra, we could solve this equation for > in terms of z and then intagrate
the result to get v which then determines y =e” as a function of 2. ‘The
algebraic solution for = is messy and we will not be able to find a clesed form
expression for the antidervative v, so we will abandon the calculation at this
point. In practical applications one would generally use power series techniques
‘on oquations of this form and ealeulato as many torms of the Taylor sorios of
the solution as are needed to give the desired accuracy. °
Next consider equations of the form
(cr + yy +er)de+ (aor + boy +e) dy = 0.
Ife = ¢2 = 0, the equation is homogeneous of degree 1. If not, try Jetting
@=2—hand gy —y—k. Wo try to choose h and k to make the equation
dz and dj = dy.
homogeneous. Since A and k are constants, we have dz
‘Then our equation becomes
(aye + ah brut bik + cy) d+ (age + anh + bot + bak + c2) dy) = 0.
We want ayh +bik = =e and agh+ bak = —c>, We can always solve for h and
unless
a |
a b| 24CHAPTER 2. FIRST ORDER ORDINARY DIFFERENTIAL EQUATIONS
So suppese
a by
a2 bs
0.
‘Then (03,b:) = m (a,b). Let
by £0, we have
aye + byy. Thon de = a de + by dy. IE
dz~aydr
Br
mete
Pra bath
‘This is a separable equation.
Ie by =0 but by #0, we use 2 = ag + bay instead. Finally, if both 6; = 0
and b; =0, then the original oquation is separable. Chapter 3
Applications and Examples
of First Order Ordinary
Differential Equations
3.1 Orthogonal ‘Trajectories
An equation of the form f(2,y,C) = 0 determines a family of curves, one for
every value of the constant C.
3.1 shows curves with several values of
C. Differentiating f(2.y.C) = 0 with respect to x gives a second equation
aN
Ste
a
|
Figure 3.1: Graphs of 2? + y — C= 0 with various values of C. 26CHAPTER 3. APPLICATIONS AND EXAMPLES OP FIRST ORDER ODE'S
o(z.u,C.dy/dz) = 0, which is explicitly given by
of of dy
9> 5 BO) + 5, (estO) FE
Example 3.1. Wo havo
3 dy
Pes 4Cet=oimplin ae ea +4009 =0
We can eliminate C between the equations J = 0 and y = 0, that is,
‘Therefore,
‘This yields a first order DE, where the solution includes the initial family of
curves. *
So just as a first order DE usually determines a one-parameter family of
curves (its solutions), @ one-parameter family of curves determines a De.
A coordinate system (Cartesian eoordinates, polar coordinates, ete.) consists
of two families of curves which always intersect at right angles. ‘Two families of
curves which always intersect at right angles are called orthogonal trajectories
of cacl other. Given a one-parameter family of curves, we wish to find another
family of curves which always intersects it at right angles.
Example 3.2. Find orthogonal trajectories of y = C sin(2). *
Solution, Figure 3.2 shows the plot of y = Csin(z) with several values of C.
We have © = w/sin(e), 0 it follows that
dy sate)
ae aia
vem)
= yoot(a),
Fin( 2)
‘Two curves meet at right angles if the product of their slopes is ~1,1.€., Maw =
=1/mous. So orthogonal trajectories satisfy the equation
dy 1 tan(z)
uv 3.2, EXPONENTIAL GROWTH AND DECAY 27
Figure 3.2: A plot of the family y = Csin(x) (in black) and its orthogonal
trajectories (in gray).
‘This is « separable equation, and we quickly have
vdy~= taste)
£ cimleasto +c,
y=4VTmleoO FC
Cis sso 'an arbitzary constant. °
3.2 Exponential Growth and Decay
‘There are many situations where the rate of change of some quantity a is pro-
portional to the amount of that quantity, L., dr/ét = kx for some constant &.
‘The general solution is x= Ae
1. Redium gredually chonges into uranium. If x(4) roprosents the amount of
radium present at time ¢, then dr/dt = ka. In this case, k <0.
2. Bank interest. The amount of interest you receive on your bank account is
proportional to your balance. In this esse, k > 0. Actually, you do not do
quite this well at the bank because they compound only daily rather than
continuously, eg, if we measure this time in years, our model predicis
that x = age, where ao is your initial investment and & is the interest
wu(iet. 28CHAPTER 3. APPLICATIONS AND
EXAMPLES OF FIRST ORDER ODE’S
whore n is the number of interest periods throughout the year. Typically
n= 865 (daily interest), but the difference is small when n = 365. Note
5 (1+ kt/n)" = eft
that lim.
3. Population grouth of rabbits. ‘The number of rabbits bora at any time
is roughly proportional to the number of rabbits present. I 2(1) is the
number of rabbits at time ¢, thon di /de = kx with k > 0. Obviousiy, this
model is not accurate as it ignores deaths.
Example 3.3. The halflife of radium is 1600 years, ie., it takes 1600 years for
dualf of any quantity to decay. If 2 sample initially contains 50, Low long will
it be until it contains 45 «2 *
Solution. Lot 2(t) be the amount of radium present at time # in years. ‘Thon
we know that da/dt = ka, s0 x = xye!". With 2(0) = 60, we quickly have
© = 50e™. Solving for ¢ gives In(2/50) /k. With r(1600) = 25, we have
25 = Hel", “Theretore,
1
200k = n($) =-10(2),
giving us k= —1n(2)/1000, When 2 = 45, we have
tu2/00) _ _1n(45/50)_ 599 In(8/10) o/s)
Stu fon ~~ yay = 1° “rey
v9 eles
1600-08 1e00 x 0192 2 2482
‘Therefore, it will he approximately 2432 years until the sample eomtains 45 g
of radium. o
3.3. Population Growth
Earlier, we discussed population growth where we cousidered only births, We
now consider deaths also. Take, as our model, the following
Let N(t) be the number of people at time t. Assume that the land is inteinsi
cally capable of supporting 1 people and that the rate of increase is proportional
to both Nand LN. Then
aN
SH EN(L-N) 9.4, PREDATOR-PREY MODELS 29
and the solution is i
TEEN
where k is the proportionality constant.
3.4 Predator-Prey Models
Example 3.4 (Predator-Prey). Consider land popniated by foxes and rab-
bits, whore the foxes prey upon the rabbits. Let 2(¢) and y(t) be the number
at time ¢, In the absence of predators, at any
time, the number of rabbits would grow at a rate proportional to the number
of rabbits at that time, However, the presence of predators also cauises the
number of rubbits to decline in proportion to the number of encounters between
fa fox and a rabbit, which is proportional to the product 2(¢)y/(#). Therefore,
da /dt = ax ~ bey for some positive coustants a and b. For the faxes, the
presence of other foxes represents competition for food, so the number declines
proportionally to the number of faxes but grows proportionally to the number
of encounters. Thorofore dy/dé = ex + dry for some positive constants ¢ and
d. The systen
of rabbits and foxes, respective
= ar — bry,
~cy + dry
BIS 815
4s our mathematical model.
If we want to Sud the function y(ic), which gives the way that the number
of foxes and rabbits are related, we begin by dividing to get the differential
equation
dy _ ~cy tary
de ~ r= bey
with a,b
£(t),u(t) positive.
‘This equation is separable and can bo rewritten as
(ot) dy _
u
tae) dx
ng gives
aln(y) ty = —eln(z) + de + 0, S0CHAPTER 3. APPLICATIONS AND
EXAMPLES OF FIRST ORDER ODE’S
or equivalently
yet = kere Gl)
for some positive constant k = eC
‘The graph of a typical solution is shown in Figure 3.3.
4,h= 1.58,
ng at a point such as a, where there are few rabbits andl fow foxes, the
fox population does not initially increase much due to the lacie of food, but with
se few predators, the number of rabbits multiplies rapidly. After a while, the
point bis reached. at which time the large food supply causes the rapid increase
in the number of foxes, which in turn curtails the growth of the rabbits. By
the time point cis reached, the large number of prodators causes the nuanber of
rabbits to doerease. Eventually, point d is reached, where the number of rabbits
las dedined to the poiut where the lad of food causes the fox population to
decrease, eventually returning the situation to point a. *
3.5 Newton’s Law of Cooling
Lot T and T, be the temperature of an object and its surroundings, respectively
Let Ty and Ty, be initial temperatures. Then Newton's Lew of Cooling states 3.6. WATER TANKS at
that
a
BeKT-1, k>0
As T changes, the object gives oF takes heat from its surroundings. We now
have two eases
CASE 1: T, is constant. ‘This occurs either because the heat given off is
transported elsewhere or because the surroundings are so large that the eontri-
bution is nogligible. In euch a case, wo have
ar
Gt eT=an,
which is linear, and the solation is,
Toe +7, (1)
Case 2: The system is closed. All heat lost by the abject is gainod by its
surroundings. We need to find TT, as a function of T. We have
heat gained
change in temperature = ae
weight» sp
Therefore,
(1g)
<2).
‘This is linear with different constants than the previous case. The solution is
hoe?) (1-eMsts)
3.6 Water Tanks
Fre
Example 8.5. A tank contains a salt water solution consisting initially of 20 ke
of salt dissolved into 10 ¢ of water. Fresh water is being poured into the tank
at a rate of 3'/min and the solution (kept uniform by stirring) is flowing ont at
2Y/min, Figure 34 shows this setup, Find the amount of salt in the tank after
5 minutes. + S2CHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
3 lite/ mim
{LE
lr
2 itre/ min,
Figure 34: Fresh water is being poured into the tuk as the welhunixed solution
is flowing out.
Solution. Let Q(t) be the amount (in kilograms) of salt in the tank at time t
(in minutes). ‘The volume of water at time t is
1043/2 = 1048
‘The concentration at timo t is givon by
volume 1040
amount of salt _Q
kg, per litre. Then
~ (rate at which salt is leaving)
de
iO+e tore
‘Thus, the solution to the problem is the solution of
ta
dt ~i0+t
evaluated at c that it is simply a separable equation,
7 3.6. WATER TANKS 33,
‘To solve it, we have
aQ_ 4 #
Qo ne
[e-3fav
In(lQ)
In(iQ) =In(0+ 4) +€,
1a(|L0 + ¢)),
where C isn constant, ‘Thus,
In(JQ)) =In(j10 +4) *) +0
=tn(Ajio+ 4),
i@\=4ho+d?
where A =e. But Q > 0 (we cannot have a negative amount of salt) and ¢ > 0
(we do not visit the past), so we remove absolute value signs, giving us
Q=A(0+t)?
Initially, ie., at ¢= 0, we know that the tank contains 20kg of salt. Thus,
the initial condition is Q(0) = 20, and we have
Q(20)
_A i»
> toy
Our equation therefore becomes
2000
“i
8) Foe?
Figuro 3.5 shows a plot of the solution. Eval
2000
1
Q(5)
‘Therefore, after 5 minutes, the tank will contain approximately 8.89 ky of salt.
‘Additional conditions desired of the solution (Q(0) = 20) in the above exam-
ple) are called boundary conditions and a differential equations together with
boundary conditions is ealled a Loundary- value problem (BVP). SACHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
~
Figure 8.5: The plot of the amount of salt Q(t) in the tank at time f shows that
sal leaves slower as time moves on.
A= pipe i the general solution to the DE 4 = — #6. Q = (0, inthe
solution to the boundary value problem “2 = —2#.; Q(0) = 200. Boundary-
value problems like this one where the boundary conditions are initial values of
various quantities are also called initial-value problems (IVP).
3.7. Motion of Objects Falling Under Gravity
with Air Resistance
Lat 2(t) be the height at time f, meusured positively on the downward direction,
If-we consider only gravity, then
a
ae
is a constant, denoted 9, the acceleration due to gravity. Note that F = ma =
mg. Air resistance encountered depends on the shape of the object and other
things, but under most circumstances, the most significant effect is a foree op-
posing the motion which is proportional to @ power of the velocity: So
B= mg— kv"
and
(a) 3.7, MOTION OF OBJECTS FALLING UNDER GRAVITY WITH AIR RESISTANCE35,
which is a second order Dg, but there is no 2 term, So it is first order in 2’.
Therefore,
a kon
aoa
‘This is not easy to solve, so we will make the simplifying approximation that
= 1 (if vis small, ther is not much difference between 2 and v"), Therefore,
we have
‘The integrating factor is
Therefore,
Note that
fo foot
cae
Thus, we finally have S6CHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
3.8 Escape Velocity
Let 2(t) be the height of an object at time ¢ measured positively upward (from
the centre of the earth). Newton's Law of Gravitation states that
kmM
whore m is the mass of the object, IM is the mass of the Earth, and k > 0. Note
that
vat
We define the constant g known as the "acceleration dus to gravity” by 2!” = —g
when x = R, where R= radius of the earth. Sok = gf?/M. Therefore
2” = ~gR®/2”. Since t does uot appear, letting w= de jdt so that
@x_dede dy
«
will reduce it to first order, Thus,
for some arbitrary constant C. Note thet v decreases as x increases, and if vis
not sufficiently lange, eventually v = 0 at soa height rpq- SOC = —9R? [tae
i vat)
Suppose we leave the surface of the Earth starting with velocity V. How far 3.9. PLANETARY MOTION 37.
will we get? (v= V when x= it). We have
‘That is, if V? < 2g, we will got as far as
2g
2Rg— V?
and then fall back. ‘To escape, wo nocd V > yIRG. Thus, the escape velocity
is VIM ~ 6.9%, or roughly 25000 mph.
3.9 Planetary Motion
Newton's law of gravitation states that
F
a
Suppose we let d= —k/r? and ag =0. Then
a_sin(@) + a, e05(8).
a, = a c0s(@) +a, sin(8), ay =
Let ¢ =reos(@) and y= rsin(@). Then
af =r! cos(@) ~ rsin(@)0",
a, =a" = 1" cox(@) — 1 sin(9)0! — 1 sin(@)6" — r cos(@) (6")? — rsin(@)9",
af =r" sin(6) + recs(O)0",
yf" =r" sin(@) +r’ cos(8)6" +r! ccs( 6)" — rsin(@) 6")? + rcos(a)o”
= 1""sin(0) + 2/8’ cos(0) — (0')" rsin(O) + 6"rcos(@). S8CHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
Now ay becomes
a, = 1” cos? (6) — 26” sin(6) cos(0) — (0)” rcos?(@) — 0” sin(@) cos(@)
41" sin!®(0) + 2r'0" sin() cos(9) — (0)? rsin?(0) + 0rsin(0) cos(6)
a Oyr
‘and ay becomes
aay =~" sin(@) con(0) + 216 sin2(0) + (0)? rsin(0) cos(0) + Or sin?(O)
+1!” sin(@) cos(0) + 2r'8' eox?(B) — (01)? rsin(@) con(6) +” rens?(A)
= 2r'6' +0".
Note that
Ore dS, @)
216 +6" = 0. bes)
Eqnation (e*) implies that”
Der! 4 0 7? = 0 => 20! =
oa
Tete
We want the path of the planet, so want an equation involving rand @, ‘There-
foro, we oliainate t between Equation (+) and +20’ =k. Thus,
@rhh _sdrhadrh
sof [Orman [2 a,
s0 44/8 = r(0)2/2, So imonrease, A’ = 1/2. Therefor,
[Ea (arn sept out is proportional to tne). 3.10. PARTICLE MOVING ON A CURVE 39
Let u=1/r. Then
du Ld
do ~ do"
@u_ 2 (dr? Adee
‘Therefore, Equation (+) implies that
ae (4) r= -
we \2
a,aeu _ a2 2
ate we = hut, uA 0,
eu
de
Therefore,
k
WL = Boon 8) 4 5
where
Therefore,
Faerie crea =
A
1- Cir - Cor,
a +y*) =(1- Cir — Cay)",
we
which is the equation of an ellips»
3.10 Particle Moving on a Curve
Let x(1) aud y(t) denote the x and y coordinates of a point at Lime ¢, respectively.
Pick a reference point on the curve and let s(t) be the arc length of the pice
of tho curve between the reference point and the particle, Figure 3.6 illustrates
this idea, Then 40CHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
Reference point
Particle
Figure 3.6: A particle moving on a eurve,
@s
a
‘The vector v is in the tangent dizection, The goal is to fiud components a in
the tangential and normal directions, Let @ be the angle between the tangent
Fine and the = axis. Then
= 2, sino -
cos(@) = FE, sin(o) ~
In the standard basis, we have a = («,4y), where
a= Ge w=
‘To get components in T.n basis, apply the matrix representing the changp of
basis from T_n to the standard basis, ic. rotation by ~8, ic.
[ ©03(6) we l[E]-[ a, c05(0) + ay sin(@)
~sin(@) cas(9) | | a, ay sin(é) + a, c0s(0)
ox = arcon(9) +aysin(@) = aE + 0,
ag ANE
oa a) as
ay = ~ag sin(6) + ay cos(0) =
and
Also, we have v = \/iE-+ #3. $0
Pa do _ Doo 20% _ vate + yay _
de dt = 3.10. PARTICLE MOVIN
ON A CURVE a
Bat this was clear to start with because itis the component of a in the tangential
direction that affects ds/dt
We now have
dy _ dyjdt
ae ~ ae fae
ns GE — WE
2
$0 that aq = v2/R. ‘The quantity Mis called the radius of curvature. It is the
radius of the circle which mostly closely fits the eurve at that point. ‘This is
shown in Figuro 3.7. If the eurve is « cirele, then Ris a constant equal to the
Figure 3.7: The radius of curvature is the cirele whose edge best matehes a
given curve at a particular point.
radivs
Conversely, suppose that 2 is a constant, Then
Ry = (14) 42CHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
Let z= a. Then
and
‘Therefore,
ve
4h? -(2- A+B,
(By = BP (2-4),
— A)? + (y— By = PP.
Suppose now dat the ouly external force F acting ou tbe particle is gravity,
Le, F = (0,—mg). Thea in terms of the T, m basis, the components of F are
cox(#)sinf) ][ 0 ] | ragsina) ] _ [ -mod#
—sin(@) eos(@) | | —mg | ~ | ~mgeos(6) | | —mg%# 3.10. PARTICLE MOVING ON A CURVE 43,
Equating the tangential components gives us
Bs
MR
‘o
-a(u~ uo) = 9(0h ~ 0),
2a = 9).
‘To proceed, we niust know y(s), which depends on the curve. Therefore,
ds
Palo — ue)
twa f et __
TT
Ifthe inaq ever becomes greater than the normal component of P, the particle
will fly off the curve, This will happen when
dt=+
ag
-%
de deda_ dr
aa a"
Theraore,
—gay (BY a=
we (@) (ds/ts) | ayy) =14 (07
yr _ yl —3)
a+r 1+?
Example 3.6. A particle is placed at the point (1,1) on the curve y = 2° and
released. It slides down the curve under the influence of gravity, Determine
whether or not it will ever fly off the curve, and if so, where * AACHAPTER 3. APPLICATIONS AND EXAMPLES OF FIRST ORDER ODE’S
Solution. First note that
ven, wah
Therelore,
"(v- wo) =1- (a),
120 (e* = 1) =1 4924,
1224 — 192 = 14924,
Set 122-1=0. 4
Solving for the relevant value of x gives us x ~ —0.083, which corresponds to y =
0.00057. Therefore, the particle will ly off the curve at (0.088, ~0.00057
Figure 3.8 shows the graph.
Figure 38: A particle oleased from (1,1), under the infinonce of gravity, will
fly off the carve at approximately (0088, ~0.00057). Chapter 4
Linear Differential
Equations
Although we have dealt with a Jot of manageable eases, first order equations
are difficult in general. But second order equations are much worse. ‘The most
manegoable case is linear equations. We begin with the general theory of linoar
differential equations. Specific tecliniques for solving hese equations will be
siven later.
Definition (nth order linear differential equation)
Au nth order linear diferential equation is au equation of the form*
2) 4 pe) 4 + Pe)
ie f
a +Palelv = te).
@
ar
‘An nth order linear equation can be wrltten as a linear system (see Chapter
9) as follows. Let
dy oly
(x)= wx), ante) =, ule) = 4
Thee dus , dita
Ls de
“If he colficen of d"y/de™ iso 1, we can divide the equation through this exefBicient
to-write im normal form. A linear eqation is in normal form whon the coocient of of) is 46 CHAPTER 4. LINEAR DIFFERENTIAL EQUATIONS
= Pe) EY Px + QC
= —Pu(s)yn ~~~ Palen + Qa)
‘Therefore, a
®
