Richardson Extrapolation Method-Numerical Analysis-Lecture Handouts, Lecture notes of Mathematical Methods for Numerical Analysis and Optimization

This course contains solution of non linear equations and linear system of equations, approximation of eigen values, interpolation and polynomial approximation, numerical differentiation, integration, numerical solution of ordinary differential equations. This lecture includes: Richardson, Extrapolation, Method, Accuracy, Derivative, Function, Manner, Two, Point, Truncation, Error, Taylor, Series

Typology: Lecture notes

2011/2012

Uploaded on 08/05/2012

saruy
saruy 🇮🇳

4.5

(120)

130 documents

1 / 3

Toggle sidebar

This page cannot be seen from the preview

Don't miss anything!

bg1
Numerical Analysis –MTH603 VU
© Copyright Virtual University of Pakistan 1
R
RI
IC
CH
HA
AR
RD
DS
SO
ON
N
S
S
E
EX
XT
TR
RA
AP
PO
OL
LA
AT
TI
IO
ON
N
M
ME
ET
TH
HO
OD
D
To improve the accuracy of the derivative of a function, which is computed by starting
with an arbitrarily selected value of h, Richardson’s extrapolation method is often
employed in practice, in the following manner:
S
Su
up
pp
po
os
se
e
w
we
e
u
us
se
e
t
tw
wo
o-
-p
po
oi
in
nt
t
f
fo
or
rm
mu
ul
la
a
t
to
o
c
co
om
mp
pu
ut
te
e
t
th
he
e
d
de
er
ri
iv
va
at
ti
iv
ve
e
o
of
f
a
a
f
fu
un
nc
ct
ti
io
on
n,
,
t
th
he
en
n
w
we
e
h
ha
av
ve
e
()()
() 2
( )
T
T
yx h yx h
yx E
h
Fh E
+
−−
=
+
=+
W
Wh
he
er
re
e
E
ET
T
i
is
s
t
th
he
e
t
tr
ru
un
nc
ca
at
ti
io
on
n
e
er
rr
ro
or
r.
.
U
Us
si
in
ng
g
T
Ta
ay
yl
lo
or
r
s
s
s
se
er
ri
ie
es
s
e
ex
xp
pa
an
ns
si
io
on
n,
,
w
we
e
c
ca
an
n
s
se
ee
e
t
th
ha
at
t
246
123T
E chchch
=
+++
T
Th
he
e
i
id
de
ea
a
o
of
f
R
Ri
ic
ch
ha
ar
rd
ds
so
on
n
s
s
e
ex
xt
tr
ra
ap
po
ol
la
at
ti
io
on
n
i
is
s
t
to
o
c
co
om
mb
bi
in
ne
e
t
tw
wo
o
c
co
om
mp
pu
ut
te
ed
d
v
va
al
lu
ue
es
s
o
of
f
()yx
u
us
si
in
ng
g
t
th
he
e
s
sa
am
me
e
m
me
et
th
ho
od
d
b
bu
ut
t
w
wi
it
th
h
t
tw
wo
o
d
di
if
ff
fe
er
re
en
nt
t
s
st
te
ep
p
s
si
iz
ze
es
s
u
us
su
ua
al
ll
ly
y
h
h
a
an
nd
d
h
h/
/2
2
t
to
o
y
yi
ie
el
ld
d
a
a
h
hi
ig
gh
he
er
r
o
or
rd
de
er
r
m
me
et
th
ho
od
d.
.
T
Th
hu
us
s,
,
w
we
e
h
ha
av
ve
e
24
12
() ()yx Fh ch ch
=+++
And
24
12
() 2416
hhh
yx F c c

=+++


H
He
er
re
e,
,
c
ci
i
a
ar
re
e
c
co
on
ns
st
ta
an
nt
ts
s,
,
i
in
nd
de
ep
pe
en
nd
de
en
nt
t
o
of
f
h
h,
,
a
an
nd
d
F
F(
(h
h)
)
a
an
nd
d
F
F(
(h
h/
/2
2)
)
r
re
ep
pr
re
es
se
en
nt
t
a
ap
pp
pr
ro
ox
xi
im
ma
at
te
e
v
va
al
lu
ue
es
s
o
of
f
d
de
er
ri
iv
va
at
ti
iv
ve
es
s.
.
E
El
li
im
mi
in
na
at
ti
in
ng
g
c
c1
1
f
fr
ro
om
m
t
th
he
e
a
ab
bo
ov
ve
e
p
pa
ai
ir
r
o
of
f
e
eq
qu
ua
at
ti
io
on
ns
s,
,
w
we
e
g
ge
et
t
46
1
4()
2
() ( )
3
h
FFh
yx dh Oh



=++
N
No
ow
w
a
as
ss
su
um
mi
in
ng
g
t
th
ha
at
t
1
4()
2
23
h
FFh
h
F


 
=


E
Eq
qu
ua
at
ti
io
on
n
f
fo
or
r
y
y
(
(x
x)
)
a
ab
bo
ov
ve
e
r
re
ed
du
uc
ce
es
s
t
to
o
46
11
() ( )
2
h
yx F dh Oh

=++


T
Th
hu
us
s,
,
w
we
e
h
ha
av
ve
e
o
ob
bt
ta
ai
in
ne
ed
d
a
a
f
fo
ou
ur
rt
th
h-
-o
or
rd
de
er
r
a
ac
cc
cu
ur
ra
at
te
e
d
di
if
ff
fe
er
re
en
nt
ti
ia
at
ti
io
on
n
f
fo
or
rm
mu
ul
la
a
b
by
y
c
co
om
mb
bi
in
ni
in
ng
g
t
tw
wo
o
r
re
es
su
ul
lt
ts
s
w
wh
hi
ic
ch
h
a
ar
re
e
o
of
f
s
se
ec
co
on
nd
d-
-o
or
rd
de
er
r
a
ac
cc
cu
ur
ra
at
te
e.
.
N
No
ow
w,
,
r
re
ep
pe
ea
at
ti
in
ng
g
t
th
he
e
a
ab
bo
ov
ve
e
a
ar
rg
gu
um
me
en
nt
t,
,
w
we
e
h
ha
av
ve
e
46
11
() ( )
2
h
yx F dh Oh

=++


4
6
1
1
() ( )
416
dh
h
yx F Oh

=++


E
El
li
im
mi
in
na
at
ti
in
ng
g
d
d1
1
f
fr
ro
om
m
t
th
he
e
a
ab
bo
ov
ve
e
p
pa
ai
ir
r
o
of
f
e
eq
qu
ua
at
ti
io
on
ns
s,
,
w
we
e
g
ge
et
t
a
a
b
be
et
tt
te
er
r
a
ap
pp
pr
ro
ox
xi
im
ma
at
ti
io
on
n
a
as
s
6
2
() ( )
4
h
yx F Oh

=+


docsity.com
pf3

Partial preview of the text

Download Richardson Extrapolation Method-Numerical Analysis-Lecture Handouts and more Lecture notes Mathematical Methods for Numerical Analysis and Optimization in PDF only on Docsity!

RIRICCHHAARRDDSSOONN’’SS EEXXTTRRAAPPOOLLAATTIIOONN MMEETTHHOODD

To improve the accuracy of the derivative of a function, which is computed by starting

with an arbitrarily selected value of h, Richardson’s extrapolation method is often

employed in practice, in the following manner:

SSuuppppoossee wwee uussee ttwwoo--ppooiinntt ffoorrmmuullaa ttoo ccoommppuuttee tthhee ddeerriivvaattiivvee ooff aa ffuunnccttiioonn,, tthheenn wwee hhaavvee

( ) ( ) ( ) 2

( )

T

T

y x h y x h y x E h

F h E

WWhheerree EE T T

iiss tthhee ttrruunnccaattiioonn eerrrroorr.. UUssiinngg TTaayylloorr’’ss sseerriieess eexxppaannssiioonn,, wwee ccaann sseeee tthhaatt

2 4 6 ET = c h 1 + c h 2 + c h 3 +"

TThhee iiddeeaa ooff RRiicchhaarrddssoonn’’ss eexxttrraappoollaattiioonn iiss ttoo ccoommbbiinnee ttwwoo ccoommppuutteedd vvaalluueess ooff y ′( ) x

uussiinngg tthhee ssaammee mmeetthhoodd bbuutt wwiitthh ttwwoo ddiiffffeerreenntt sstteepp ssiizzeess uussuuaallllyy hh aanndd hh//22 ttoo yyiieelldd aa

hhiigghheerr oorrddeerr mmeetthhoodd.. TThhuuss,, wwee hhaavvee

2 4 1 2 y ′( ) x = F h ( ) + c h + c h +"

And

2 4

( ) (^1 ) 2 4 16

h h h y x F c c

HHeerree,, cc i i

araree ccoonnssttaannttss,, iinnddeeppeennddeenntt ooff hh,, aanndd FF((hh)) aanndd FF((hh//22)) rreepprreesseenntt aapppprrooxxiimmaattee

vvaalluueess ooff ddeerriivvaattiivveess.. EElliimmiinnaattiinngg cc (^1 )

frfroomm tthhee aabboovvee ppaaiirr ooff eeqquuaattiioonnss,, wwee ggeett

4 6 1

h F F h

y x d h O h

NNooww aassssuummiinngg tthhaatt

1

h F F h h F

EEqquuaattiioonn ffoorr yy’’((xx)) aabboovvee rreedduucceess ttoo

4 6 ( ) 1 1 ( ) 2

h y x F d h O h

TThhuuss,, wwee hhaavvee oobbttaaiinneedd aa ffoouurrtthh--oorrddeerr aaccccuurraattee ddiiffffeerreennttiiaattiioonn ffoorrmmuullaa bbyy ccoommbbiinniinngg ttwwoo

rreessuullttss wwhhiicchh aarree ooff sseeccoonndd--oorrddeerr aaccccuurraattee.. NNooww,, rreeppeeaattiinngg tthhee aabboovvee aarrgguummeenntt,, wwee hhaavvee

4 6 ( ) 1 1 ( ) 2

h y x F d h O h

4 1 6 ( ) 1 ( ) 4 16

h d h y x F O h

EElliimmiinnaattiinngg dd 11

ffrroomm tthhee aabboovvee ppaaiirr ooff eeqquuaattiioonnss,, wwee ggeett aa bbeetttteerr aapppprrooxxiimmaattiioonn aass

6 ( ) 2 ( ) 4

h y x F O h

  docsity.com

WWhhiicchh iiss ooff ssiixxtthh--oorrddeerr aaccccuurraattee,, wwhheerree

2 1 2 1

(^2 )

h h F F h F

This extrapolation process can be repeated further until the required accuracy is achieved,

which is called an extrapolation to the limit. Therefore the equation for F 2

above can be

generalized as

m m m m m

m (^) m m

h h F F h F

m

− − −

WWhheerree^ FF 00

((hh)) == FF((hh))..

TToo iilllluussttrraattee tthhiiss pprroocceedduurree,, wwee ccoonnssiiddeerr tthhee ffoolllloowwiinngg eexxaammppllee..

ExExaammppllee:: UUssiinngg tthhee RRiicchhaarrddssoonn’’ss eexxttrraappoollaattiioonn lliimmiitt,, ffiinndd yy’’((00..0 05 5)) ttoo tthhee ffuunnccttiioonn

yy == --11//xx,, wwiitthh hh == 00..0^011228 8,, 00..0^000664 4,, 00..0^000332 2..

SoSolluuttiioonn

TToo ssttaarrtt wwiitthh,, wwee ttaakkee,, hh == 00..0^011228 8,, tthheenn ccoommppuuttee FF ((hh)) aass

y x h y x h F h h

SSiimmiillaarrllyy,, FF((hh//22)) == 4 4006 6..6 66622773 3.. TThheerreeffoorree,, uussiinngg EEqq.. ((77..3 30 0)),, wwee ggeett

1

h h F F h

WWhhiicchh iiss aaccccuurraattee ttoo OO((hh

)).. HHaallvviinngg tthhee sstteepp ssiizzee ffuurrtthheerr,, wwee ccoommppuuttee

2

h F

And

docsity.com