0% found this document useful (0 votes)
31 views21 pages

Chapter 2 Solutions of Equations in One Variable

This document discusses numerical methods for finding solutions or approximations of functions. It focuses on the bisection method, describing how to construct a sequence of intervals that contain a solution by repeatedly bisecting an interval and eliminating halves based on the sign of the function values. It also discusses stopping criteria for the bisection method, including achieving a desired accuracy or when the interval width becomes less than a specified tolerance.

Uploaded by

Charbel Rahme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
31 views21 pages

Chapter 2 Solutions of Equations in One Variable

This document discusses numerical methods for finding solutions or approximations of functions. It focuses on the bisection method, describing how to construct a sequence of intervals that contain a solution by repeatedly bisecting an interval and eliminating halves based on the sign of the function values. It also discusses stopping criteria for the bisection method, including achieving a desired accuracy or when the interval width becomes less than a specified tolerance.

Uploaded by

Charbel Rahme
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 21

CHAPTER Soludions Eqotons in a.

Recal: calles Solution, noot, 3e of a


fanction
doubl

And n tripe
f(«) =
(*-P) k(*)
rator
e be t arees Poifive uger for hich t): P)
s
W)
M
called 4a wudhipeicity of
Pis Smpl wo f)--')e
M:2 S double oor ( )= x'()
M
noor
+)
In erad, wt Sau7 oo wR mudtipléty m
We sha assume a r P is
We shale assOme
Simpla oor of t
har p isnigu na,o)
Sopose Ainuusf
3 ¬
)f(6)'s.4.
<o
p (,b) f(p) = o

5isetion Meffod
Tdea Constrock a
Sequn P ). ofopprox. to P a tolloos
Constroc a nested
Sequwca ofsobintrvals
L.6. [a,5] 2la,, b,). 2 [ab ) 2.. 3
f(a)f(b) <o CCrtoard in we ge ab by tias aud eror
Le a ,
midpoin
Cuatk f f(«)f(P) Co, t
PE
fa)f(?) =O b
ta P. s oor don

P¬b) ) AappDximation
TO
In Cases <o ana o We REPEAT
We expecr thar in te end im p»: P
We nou psove

en p-Po<Pan bn-an
No b a b-a) bn On
ba,.)
D3-a (6a)
bn-an-)
LHS RHS CaucEqual tecmns
bo an b- a,) a

eIp-Pal - am) = -a)


eIp b-a))
im hd for esm
BSanduwich tRaoa im
en

im en 0 Simp-pal o i m Po P

cts on Ca,b, f)f(b) <o, simple nae of f in (ab)


6isection mettod qerato
svch fe
a Seguaua (P of appsDKimations
P 2

3 4- Ps 0sing biseckion efod f)= vx -


Angies in RADIANS
Cosx

f(o)f)(-J[-cost
20

o.S

f(a) f() : f(o)fo.s) (-]No.5 - Cox .S ]


[-1J Lo.4 -0.3] >>o
e
(P b) =
(o.5, ) , be)
2

f(afea)= f(o.s )f(o.1s) [»-o)[oa-co1o.35 )


- o ) (o.s6 - o.*1) ce

p (a,P) (os,o.4S) a,,b5)


Pa tby o.S0.* o .1S
2 2
rerations would qparatte on opprokimotion ? ofP
How wany
o within

aftr t dacimal
0 6 ees
o
We kow tRar
2

dhoos Such tear


2" lo

n2 alo =6nlo
6 lo =19.93
n iaer

T f uwe champe foing wil dhamge SinCa eindapeuduh of f


only dußendiur an a aud b.

12 Find a
app oxatian or 3 to witin lo osina biseckion me-fod.

3 : .432oSo 8 6 ..Exac ualu


Le f(): * 3 0 based an Re

We Kuow tear
p- <
choose n Soch tRak

n a Anto= 4 Oulo
n t o = |3.28
n 2

4
3 25S o Bisetion Meteod
Let t uürva be (2, 3)

ET
lo3
C,47 n?
L:2 te.r Vo
f(x)x*tx-4
yf)= [141-[«4-4):[-29(4*] <o

,Pa
Choose n Such ar
lo ooo

nAn > n 3000

n 3ooo l1.S5 | n > i2


C-l Pa-Pl < ba pper boundor emor

n bisehon metkod

Tn ofter nunesica metkods, we dou 'f


Stoppinq Critrion assarily have an pper bond for e r b r
When do ue Stop arating ?
Sia

Olp-P. <E Nott No is satisfactor


See ex. I6-
f()I<E
ackual eror ena
ich is miskoodn
e P P
exat approxmate

A6f6) : (x-i) P: P I

n PWe ooo

n loo

Tru
E22
loobo
00 0 *
But en : p-P <lo for nloopo

(ooo
2looo This shoos 42a
stopping oriterion (2s
RAcap o satistattory

44 (Hamou Monic Sequnu

TP. P
Houeder, P . Qim (Harmauic Harmouie Series)

Su nuwbero terms ae
diveroesto (iugl test

P.; P P1
This Shods 4a t stpp Criterion O is dnsatis taoru.
MCO
I n bisection meleod, ?a a t . Shos ak P need nor be
cAoser to ta P
2 True or Folse

P i s s r loserto p

P:a, P

Fixed oint teratiom


Poin Saud t be fired poin of funchisn
a
g )
Pis peaca ere
mees ta
bisecrorgx d:96)
Amark: . A 4c Meed not haue
aftxed poink
2 A fc ma have mohe tAan ow fired
Poin
3)
2) Find fixed poin: gk)--2x
g ) xx -intr sectian
x 2

xx-3zo
ar -3

3 A cot fnding proble f()-o A fired poin proble


f)= g)-x e e g(«) {(*)t*

xis noo off f is a tixed psnt f


Note o wRua
fG)=° k f ) =o (#o) x+ kfG)=* ZG)= *
shere ) = x* kf«)

Thm 2.3 cks on Ca,61


b )
Ca1stela at aot O fined poin Pina,b

i n oddition, exiss in a,bamd F re is o<K<I


for a E(a,s)
Such h a k)| <K
a(a)a Inpo ss.h

ProofCi ala)a or
b)- b b u a . Notaina to Pbue

Oawise,
Couside a(a g(b) <b
x
a)h(b) =
(2a)- o)(7(6)-b)<0
B UT 3 e (a,b) s. t. P)
ie P)- p.
a(P) =P P: po
) Suppose auotur xed pt of
EC,5) s e)-g(P
Cout radiction
(-
oke-q< le-q1
To opproximats P, we shol stnt wilt
we Se R
initia egpbx P aud teu
ttratim
R. () ond i P.
Conwera
P
Twn

P.P-)

Veriicad to g(x)
Ty boh cokuerge
Horieutal to
Rapur
P
Fixed Poin (Türotion) Tuoum (2.4) s u f f i u e u r conduhons

Supes crs on4,b)


b into itsef ha
maps
a0x¬(a,b) Poin
a

P in
Onipue
a1
fied
Moeover, Sqnnli Conngt toPtor an chei

Cofollary (ia
Ondar hypattesis f FPT (24). teu we have

()e lP-P. K.maxP,- ,b-p.


P'
Figed bint Ttiratian find Noo
- x - xx x - :0
oo sits in [,2)
se P

f(x) *-x -1 R uni t


natunal doia ) -X Caunor Ose tis 9
Sinu ) 2
How do we Cwoose 3Suc
3Such ar te Suffiieur Conaitdns ae
satisfed7
le start conditi5n 3 Ig()1<k<i

9):x 3 a): nx** Bad dhwl


h a s e arge
sRope
x) x

Raort 9*)=* wfe Swall derwatiwes tor gE):


x:x
x x*
Isolat +a aot Pouwer or x 4R toake hoo
Are e 3 Couditions Safisfied?
Conti oun

(iyg) sincL
Tetm
we a
workna in ,z

O.2 k<
3 V
&
amd 9 (x) > O
Alaaus incran.na
80) : 2 .2S e l1,2
9 2 ) 2):( a * : 3h I.4u e , 2

9 L] into
itsef
3 Condihons ne
satisfied ,
eare fore b FPTT
to t u
Coeng
fired tor Qn choi . é ,2J|

(.t1 2 Roden 1.26

ETC
P-P. K waxP-a, b-p.} = (o.21.wax {o,

We uaw
o.2
P Pa s(o.21)
Choose SUch ear (o.2 <1o o. ol
n Q lou) < In(o.o)

Copw
Conput yaluso ubr
fpr n
whr is
.

(-.S6) < u.6os)


Posifiue o moatiwe n .6oS .5
1S6
n 3
6 Fixed Poin iutration wt aLmac
Ose P 1
on ,2
x 3xt3 x 3 (3x3)"
Ose
Coutiuor
) ( 3 x 3 ) ° ..6x 33 3x o

EVT Find ax and n Valun for g'(«)

)) o strictly inCtanino

2) : 6" 1.S6$ ¬C,2 Rurdl3


aps ,j ixto tself.

3)"(2-?)
so-o x2
ea a)= a=0.394
-0.3911
M 4L'maxlp-a,bBz0jt8
(k=0.to3
o.to8'<*n>5l}} h247/145
in , so xFt/21)/2 ene e reel
Giugnn is guadrahe 2
routsakxE t3tD/2 ; at=13t2).ckary JP-R000210.
8- Fction; g( ): a
cafoltar42.
Ose FPT find p in [ . ia lo acuna

aG)o
na
Contimo w y : na e

G) g)- () *(): - Dna (


Stricte dacmasn@

na

nto ks

) deredaing

o 55: k
FPT
ho a
xed Poin P in
OP-P. I K . wmax a,
b-e.
K
OSin . 2
PPT Th isoution P a P ) C o v g p to p for a P.¬G
P t
o . F 0 + - .-
P
0.S5
e, lP-P. -o.55
.40o.S
o.S: o.46 (o55
Cwose Suuh

m2ad it a is decreasina
nPrv wmd hypstlasy
PoO.64u18Si
CamyBuceleulehos,ehe re
- noPEo%}|2052524."
.6+018SI`
l R-To.cvoo634olo < lot
g cearn
hAmerieranena .aciee thus ouuwaq 1 6
noudr ied
Find 3 wit acno
ans w e to tx (2 Se ction 2.
Ose FPTT
(Biscetim)
3

x, Ose ) = *wb 9():"J3

a): 3 . C,2
'(*) . o strctlu increaanG

8) : "*3 44 e [.,t) 7 L,27 ito


. &|

x)stricty dacreaAina

FPT has a u s fixed Poin


cmvn to p for
Ond t iration P

P 3p

e.p P. s o.4x
axP.a,-e.
o.S

(o.u)o <io" : 2
(oodo

Consusk unchin +Rat So.tisfien none of t h4po theois


oFPT, e ha nite fixed Poinr

FPT: G) c
mapsa,b) into ts ha a omqe fxeá
()
K< sufiieu a
Osndit.6h

s fuito doemt satsty a


of Covd t oM, 4e
fired oin
Newfon's Tan .3
f):o
lonpeuk Metkod
a Seerricad Appvach ( #s/P. Ho)
O choose P ose euouph
dauw taup + cur ve
find ta epuatitm of tRa tougg
Equution
m-
t(e) sRope
3 Find (x
P i i r secin of T, amd
X-akis
T . 0 x-axis
o- f(.) f(e.)(P. - e)
PP
f(P) P P. - ff')f(a)
PP le 4O7
9en
Rapeat
f ) #07)
ixed Point Approadh Newton as a
fixed Poia
Nota
A ( ) wk9&): x
f')
>P P. fa)
f' (e.)
Newton is speial fixed point prbl whue

)x-

Ho to ge t f)o s(x)= *

F() -o fad odlyf |f«)=o


f(*)
Rf&)eo
No nal toulas error even touph its a txed poin
poble hecn

ousader, COn
d Simp ewouh 4 find it extrewO

tRa t
OR UseCoene
of stoppn
to
eria Pbuidad SepunC P
P
4f()
: *-6 P 1 Find P Osing Newtba's

g:x6
f ' ) 2x

P. P- 6

P. : 3.5 P 3St6 8
33.5
Not 2.44 94
acuna of lo"
Ho mauaratiow would uaatu

2x 2

) -3 Inirved

Tw. Sposef e cla,b] {tuaice ctslu diff. f: la,b)-Rf


_Aists crs
f(P)-o,f'(¢)40
Th,3S >o Sud t a r Neston's weuod P. P (2)_
fP.)
Onao a Seuu for a
Initio proaiuoton

P cose
ou t
6-N
6 Neaten's
e 9+ aa ces x 6- 6 : O
cesx 1 2

f(x): e 2coix - 6 Ne Use PCI<lo

kich is onsatsfactory
Sppose If. <&. H b en -e.?

En

e eo + E
e e, t£

en eo tnE

Eror Analysis For Iürative Mettcds


(a.4)
andur of Cowuenge
P Suppose
Le
fostar eror
T +* Constas exisk s.

im r
Which is teaia?
Hous to weana Speed/ P-Pl
at/ondar of con. Than , wt Se Con
aoototic error Conshamr A
P. P Liay
d 2
P P Quadratcall
3
PaP Ciica
ETC Wik onder a

Coucusio T bigger t « , t Couentuc .

Find t Odsr of Condenpen ad te asuptotic eccbr


Cons Fau tor
a)P b)
A h -
3^ 3 3--

() 3M)
V3
2-1)-I
oX

C Ys
Lincanky
Wite ordar k:

(b) a -
(2
im

AWhat F Ao
A A P. P
e
ene O

xlaylor eries

feC()
a ¬T
for ET , ad (a,x )
for eack n, are exists e E

Tager fG) : fa), f a (x-a) + f(a(x - a - a(x-a)


+ (x-a)
nt)!
ce(a,r, n) [funetim n o Polguoid
R (a) R.() : )(-a" Ramaander Eror
(n+)
m R() : o . te
no

fo): fla) a(-a) fa) (-a). fal


n
(a-a)"
(nwr)
(a-o) n
TayeorSeis f fat
aulinad at a)

For Who x is R.(): o? Dsing Rat io Tes ar n* nos b


-
we Co find fea selutis
X
Smap x al <ë< Sad
SmaC
(n+)

f 1*""(xs M Bodud)V safe


Sinx

Tayhor's amposinq funeisn


Macdourun's Xx-a 0, we ay assuma
a :0
(x-o n

e' -a
x

e e e

Malaunin Tayor at O*SinKx G


(nt)!
Sinx uan
Sin ((x -a) * a )

n'
(n+!

R o f«) P.(«)

no f&) - f{a)+ f(e)( - a) Dotunaa bu MUT


f f(a) i
Hozea i
-Fa) )-a) fCa)
(Qaas to b.g emors)

n: f) fa) fa)(r-a) f ) (K-a)


tan ia a o

n2 f fo)t iel -a) £(a ( ) , f a - a


3

0:Tayor:
h

sip si no uh nfo at a

Jamc
Pjettl We Co
find f(arh)
f)= xsin x m:3
lan-firs
XSinkxxx- IfG):*s* Cdilfensna
t

Mus Meoi
n n)
x+h
Son x<c = C(x,h, n)

P : ooE wik ultipkiáty m 2 f f{x)f


Reeall:
3 fx) -p h(*)| (p) #o
har: Pp is oot wit& ultipliáty m m2)
Pove Hha: a

f(e) o ad f(P) #o
fCe) f'(o)- -

fo) ) - 3 ) ' ( » s )wheu ) : - ) +S)


x-3) h)
h (3) (3-)3 (35) : 2 #0 4.
Thinton is oo a of f witudtipliihy
Sans
Sam far4f u ) -1) &) hena ) -3) (x+s)
oot 0.
t(x) : x sinx Find multipli cdty of ta

Cx-o)"hC«) hlx) I -
ho)4
xsinx f(o): o
Meltod 2 eb's rnpua +(x) Sinx+ xCosx f'o) o
f(x )=
Cosx Xsink f(o) 2 #0
" (x) Cosk

Thinforu O is hoot wik kultipkiay m: 2


Thatiue Meltods
Bak to Emor Analysis of
Find Ph P fixed pt of fo a

AF whidh order

Racollkim L :A
PP
We have
R8 (P.) 9(P (2°P)
he Sndl

n (ner)

3(P) * (P)...-
.P.-P
P '

m
Ce P) P

m uadrat col
P

whicolg

B tu Newton fC)o, P noot ie f(e) = o


a iked 2oink
s) x-f ) firedoin of
HW Sppose P Sima root f f ACp) o, fe) 4o
PP -f()
f(P..
Cne adraticoll.

So
8)H2fG)
f&) 2 p
na druti cally
Cane P
9x-
ti
What the anyp totc Emr Costak ?

Now p is
Pis
a multpla hoot of 4, oife multiplc.tu m2

eton's PP
fCe.
Tule wads to acclnaa canu.

by usin
Newton's
Shpl nook
P P
nad.
fa)

Tuo ways to accalerat


Usemodifred Nes ton's meRod P P.. -m t l - )
f(?)
i) pis a
simple heot of (x) t : aud te use Neuofon's
appli'cd to
Sotution
a
P.P P-
fl?.)
- (P. -P)- P h(A)
(P p.)* (2-P)*"(P.)
(P.-P n(P.) (a-P)H(2)
P
P. -P
) We have
Pm f(
fP)
PPP. -P -m
)
F(a.)
(P.-P) -P-P) hle)_
PP)p. -ps4/e)
CP-)+()*(.) J
:(Pa- P
wACP)(?.-P) KCe.)
(P.-P
P)+(P-P) *(e.)
P.-P
a) ) f
4.T (

ohu Hx):(k-p) h(*)

AND
H()
PSimpl toot of «)
Nehon's mekod P P P madrakicol
PP CP)

PP. f'()
f.y-f(e.)F"(e.)
F'e.
eWhich metead n
b) «prackical?
Modrred Nosfon's is hpractical becowoe we dou' F

Puou
P.&b

6) P.
For an e
intiqur k e
I,z,3,.- Sep. P. 0
ok

(-) o<

3-a) P lo
en lo
2
oC«-2)
e
Clo 2
d oroder :3
A9-Construc
-3
lo

P ordr o S

-2"

R offw fipllcity

Thou 3. :P Simpe oc of f f(P) =o, f(^)#0


Modify . to sw ta Dfiphiit

e) 4o
j lit)=f(o)f"- AND
(E) Taykor fa): f{e). -p) -*+1d m.

fa) 6-h) Son c hetween


kuu ()= ( c o )
C:c()
(P) ( )
x

())
+0 Gien)
8-6) P. lo doesn't Convenp& adratica
upndes of K sige k

2 lol.(J*

You might also like