Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > comp.soft-sys.math.mathematica > #10992 > unrolled thread

Solving Third Order differential equation using Mathematica.

Started byamitesh <singh.amitesh@gmail.com>
First post2012-07-20 07:53 +0000
Last post2012-07-21 03:42 +0000
Articles 3 — 3 participants

Back to article view | Back to comp.soft-sys.math.mathematica


Contents

  Solving Third Order differential equation using Mathematica. amitesh <singh.amitesh@gmail.com> - 2012-07-20 07:53 +0000
    Re: Solving Third Order differential equation using Mathematica. "Nasser M. Abbasi" <nma@12000.org> - 2012-07-21 03:39 +0000
    Re: Solving Third Order differential equation using Mathematica. stephen.p.luttrell@gmail.com - 2012-07-21 03:42 +0000

#10992 — Solving Third Order differential equation using Mathematica.

Fromamitesh <singh.amitesh@gmail.com>
Date2012-07-20 07:53 +0000
SubjectSolving Third Order differential equation using Mathematica.
Message-ID<jub2q5$ehn$1@smc.vnet.net>
Hi,

I am trying to solve a third order differential eqn in mathematica.
DSolve[{1/30 f'''[x] - f[x] (2 a^3/(1 - x a)^3 + 1/(1 - x)^3) - 24 a/
(1 - y) (1/(1 - x)^7 - a^5/(1 - ax)^7) == 0, f[0] == 0,
   f''[0] == 0}, f[x], x]
 I am running this eqn for last 1 hour :/ why does Mathematica take so
much time?

Will Mathematica able to solve this eqn?


Thanks
Amitesh

[toc] | [next] | [standalone]


#11013

From"Nasser M. Abbasi" <nma@12000.org>
Date2012-07-21 03:39 +0000
Message-ID<jud89v$ivl$1@smc.vnet.net>
In reply to#10992
On 7/20/2012 2:53 AM, amitesh wrote:
> Hi,
>
> I am trying to solve a third order differential eqn in mathematica.
> DSolve[{1/30 f'''[x] - f[x] (2 a^3/(1 - x a)^3 + 1/(1 - x)^3) - 24 a/
> (1 - y) (1/(1 - x)^7 - a^5/(1 - ax)^7) == 0, f[0] == 0,
>     f''[0] == 0}, f[x], x]
>   I am running this eqn for last 1 hour :/ why does Mathematica take so
> much time?
>

I do not know why. It looks like a hard ODE.

But let start first by fixing few error in your code

1)  ax  should be  a*x

2)  a third order initial conditions ODE will have 3 initial conditions
(unless you wanted to obtain constants for these. But since you provided
IC for f[0] and f''[0], then this tells me you missed an IC for f'[0] ?

3) what exactly is 'y' above? this is used as a parameter like 'a'. Then
using 'y' for a parameter is a really bad choice for a symbol. Use 'b'
or 'c'. That is the standard.

4) it is a hard ode. For some conditions on 'a' and 'y', you'll can get singularity.
Try adding assumptions on the parameters to see if this helps. But
why not simply try NDSolve. Why do you have to have an analytic solution?

NDSolve tells me this is a stiff ODE as well.

--Nasser

[toc] | [prev] | [next] | [standalone]


#11016

Fromstephen.p.luttrell@gmail.com
Date2012-07-21 03:42 +0000
Message-ID<jud8fl$j1o$1@smc.vnet.net>
In reply to#10992
The differential equation can be solved for a=0, and if you series expand the differential equation in powers of a about a=0, discarding 2nd and higher powers of a, then it is still possible to solve it. I haven't investigated any further.

Here's what I did:

    DSolve[{Normal@
         Series[1/30 f'''[x] - f[x] (2 a^3/(1 - x a)^3 + 1/(1 - x)^3) - 
           24 a/(1 - y) (1/(1 - x)^7 - a^5/(1 - a x)^7), {a, 0, 1}] == 0, 
       f[0] == 0, f''[0] == 0}, f[x], x] // Simplify

I haven't quoted the output because it is quite large.

[toc] | [prev] | [standalone]


Back to top | Article view | comp.soft-sys.math.mathematica


csiph-web