Need help solving pseudo-hypergeometric ODE

  • Context: Graduate 
  • Thread starter Thread starter pbecker314
  • Start date Start date
  • Tags Tags
    Ode
Click For Summary
SUMMARY

The discussion centers on solving the pseudo-hypergeometric ordinary differential equation (ODE) given by x*(1+x)*y''(x) + (3 + A*x + B*x^2)*y'(x) + (C + 4*B*x)*y(x) = 0. Pete Becker from George Mason University seeks a closed-form solution, ideally in terms of confluent hypergeometric functions, but currently only has access to power series solutions via the Frobenius method. The Maple software indicates two independent solutions involving the HeunC function, but these are not convenient due to the lack of a global theory for Heun functions and the complexity of their computation. Despite these challenges, Pete can continue his research using the Frobenius solutions.

PREREQUISITES
  • Understanding of ordinary differential equations (ODEs)
  • Familiarity with the Frobenius method for power series solutions
  • Knowledge of confluent hypergeometric functions
  • Experience with Heun functions and their properties
NEXT STEPS
  • Research the properties and applications of Heun functions
  • Explore transformations for confluent hypergeometric functions
  • Study advanced techniques in solving ODEs, particularly using Maple
  • Investigate the global theory of Heun functions for better computational methods
USEFUL FOR

Mathematicians, researchers in applied mathematics, and anyone involved in solving complex ordinary differential equations, particularly those interested in hypergeometric and Heun functions.

pbecker314
Messages
3
Reaction score
0
Hi All;

In the course of my research, I need to solve the ODE

x*(1+x)*y''(x) + (3 + A*x + B*x^2)*y'(x) + (C + 4*B*x)*y(x) = 0

It's somewhat similar to the hypergeometric equation, but so far I can't find any closed-form solution. Of course, I can obtain power series solutions using the Frobenius method, but I would very much like to obtain a closed-form solution, even if it's only valid for specific values of the constants A, B, and C. I was thinking that a solution in terms of confluent hypergeometric functions might be possible, but I can't find the required transformation. Any help or suggestions would be appreciated.

Thanks,

Pete Becker
George Mason University
 
Physics news on Phys.org
Maple says two independent solutions are...
[tex]y \left( x \right) ={{\rm e}^{-Bx}}{\it HeunC} \left( -B,-4+A-B,2,-1/2\,B \left( -8+A-B \right) ,-1/2\,{B}^{2}+1/2\,B \left( -8+A \right) -3/2\,A+C+5,1+x \right)[/tex]
and
[tex]y \left( x \right) ={{\rm e}^{-Bx}}{\it HeunC} \left( -B,4-A+B,2,-1/2\,B \left( -8+A-B \right) ,-1/2\,{B}^{2}+1/2\,B \left( -8+A \right) -3/2\,A+C+5,1+x \right) \left( 1+x \right) ^{4-A+B}[/tex]

Not really much help, since the Heun C function is defined as solution of a certain differential equation...
 
Thanks, I do appreciate the effort! I figured there might be Heun solutions, buy they are not very convenient since there is no global theory for those functions. Also, they need to be computed using a two-term recurrence relation, which is a bit of a pain. Hence my hope for a hypergeometric-type solution, at least in some special cases. Nonetheless, I can proceed with my work based on the power series Frobenius solutions (essentially the Heun functions), so it's not a show-stopper for me, just a bit more work! Thanks again,
Cheers,
Pete
 

Similar threads

  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 28 ·
Replies
28
Views
3K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 0 ·
Replies
0
Views
4K
  • · Replies 4 ·
Replies
4
Views
3K