Scribd
Upload
461 views3 pages

Frobenius Examples

The document provides three examples of using the Frobenius method to find series solutions to ordinary differential equations near a regular singular point. The first example shows the steps to find two independent solutions as power series in B. The second example finds just the first three terms of one solution around B=0. The third example again finds a power series solution but notes the computer was unable to find a clear pattern to all the coefficients. It also finds the second solution contains a logarithmic term.

Uploaded by

poozar
Copyright
© Attribution Non-Commercial (BY-NC)
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
461 views3 pages

Frobenius Examples

The document provides three examples of using the Frobenius method to find series solutions to ordinary differential equations near a regular singular point. The first example shows the steps to find two independent solutions as power series in B. The second example finds just the first three terms of one solution around B=0. The third example again finds a power series solution but notes the computer was unable to find a clear pattern to all the coefficients. It also finds the second solution contains a logarithmic term.

Uploaded by

poozar
Copyright
© Attribution Non-Commercial (BY-NC)
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/ 3

Frobenius method examples, page 1 of 3

Examples on Frobenius method


Example 1. Solution. At B !, the equation has a regular singular point: # # # Cww B #Cw B# B "C ! By the recipe, _ C ! +8 B8< Cw ! a8 <b+8 B8<"
8! _ ww 8! _ 8!

B# Cww a#B #B# bCw aB# #B #bC !

C ! a8 <ba8 < "b+8 B8<#

Into the equation, aligning the powers: _ B# Cww ! a8 <ba8 < "b+8 B8< #BCw #B# Cw BC #BC #C
_ #

! #a8 " <b+8" B8<


8! _

! #a8 <b+8 B8<


8! _ 8" _

! +8# B8<
8# _

! #+8 B8<
8" _ 8!

! #+8" B8<

Sum up (infinite sum from 8 #, separate terms for 8 ! 8 ")


8#

! c aa8 <ba8 < "b #a8 <b #b+8 a #a8 " <b #b+8" +8 dB8< +c<a< "b+! #<+! #+! dB< ca" <b<+" #a" <b+" #<+! #+! #+" dB<" ! Collect the terms: _ ! ca8 < "ba8 < #b+8 a% #a8 <bb+8" +8# dB8< a< "ba< #b+! B< aa< "b<+" a# #<b+! bB<" !

8#

Taking < #, guaranteed to give a series solution, we get +" +! and the recursion is +8 #8+8" +b8# 8a8" It is a homogeneous problem, so we are free to choose +! ". Anticipating division by factorials, we might change the variable to ,8 +8 a8"bx

Frobenius method examples, page 2 of 3 so that the recursion becomes, after simplification, ,! " ," # ,8 #,8" ,8 The solution of the recursion is easy to find: ,8 8 1, so 8" " +8 a8"bx 8x and the solution of the ODE is _ 8# C" aBb ! B B# /B Second solution: C# C" '
8! 8x [ # C"

# #B " C# B# /B ' B% /#B .B B# /B ' B# .B B/B B / The general solution is then C -" B# /B -# B/B ______________________________________________________________ Example 2. This time a formula for the solution was not found by computer algebra software, so I will settle for a more modest task: Find the first three terms of the Frobenius solution around B ! for BCww /B C ! Solution. For easier handling of the series, write B# Cww B/B C ! Again, _ C ! +8 B8<

# .B where [ exp' B #.B B# /#B

Cw Cww BC
# ww

! a8 <b+8 B8<"
8! _ 8! _ 8! _ 8!

! a8 <ba8 < "b+8 B8<

! a8 <ba8 < "b+8 B8<#

and we take only the first three terms: B# Cww <a< "b+! B< a< "b<+" B<" a< #ba< "b+# B<# B/B C B
B# "x <"

+! B

total: ! <a< "b+! B< aa< "b<+" +! bB<" aa< #ba< "b+# +" +! bB<# Since +! ! actually we can take +! " the larger of the integer values of < is < "

B a+! B< +" B<" +# B<# b #x " " " +" "x +! B<# +# "x +" #x +! B<$
$

Frobenius method examples, page 3 of 3 The equations simplify to +! " #+" +! ! '+# +" +! ! with solution " +! " +" " +# "# # and the series starts with " C B " B# "# B$ # Remark. The computer returned some more coefficients: " #$ "*( +$ "%% +% #))! +& )'%!! and a pattern is not clear. By the formula, C# C" ' Second solution? C# C" '

" " " "" $ B" " B "# B# ' B# B "# % B .B # " " "" $ B" " B "# B# B ln B "# B ) B# # C" ln B " " B B# # The other solution does have a logarithmic term. " # .B " B# " " B "# B# #

and since the coefficient at C in the differential equation is 0, we can take [ exp ' !.B "

[ # C" w

.B

You might also like