0
$\begingroup$

So I set up my singular perturbation problem after change of variables for equation and boundary condition. But when using DSolve for the analytical solution, there's always a 1/2 in front of it. Can someone take a look at my set up and tell me what went wrong? Thanks!

singOrd1 = 2 Derivative[1][Y0][X] + (Y0^\[Prime]\[Prime])[X] == 0
bc1YOrd1 = Y0[0] == 0
solSing1 = Y0[X] /. DSolve[{singOrd1, bc1YOrd1}, Y0[X], X][[1]]

enter image description here

$\endgroup$
5
  • 1
    $\begingroup$ People here generally like users to post code as Mathematica code instead of just images or TeX, so they can copy-paste it. It makes it convenient for them and more likely you will get someone to help you. You may find the meta Q&A, How to copy code from Mathematica so it looks good on this site, helpful $\endgroup$
    – Michael E2
    Nov 20 '21 at 23:03
  • 3
    $\begingroup$ What's wrong with C[1]/2 as a coefficient? $\endgroup$
    – Michael E2
    Nov 20 '21 at 23:05
  • $\begingroup$ Where is the 1/2 come from? $\endgroup$
    – user82861
    Nov 20 '21 at 23:28
  • 1
    $\begingroup$ It's a linear homogeneous system with a homogeneous BC: If $y$ is a solution, then so is $\alpha y$ for any scalar $\alpha$. Someone else's solution may have a different constant multiple than yours, but they're both still correct. One might be able to guess how Mathematica produced a solution with a factor of $1/2$, but I'm not sure what it matters. It just reflects different, valid choices along each solution path. $\endgroup$
    – Michael E2
    Nov 21 '21 at 1:07
  • 1
    $\begingroup$ You can get rid of the $1/2$ by solSing1 /. C[1] -> 2 C[1], if desired. It's a valid transformation, since C[1] is an arbitrary scalar constant. $\endgroup$
    – Michael E2
    Nov 21 '21 at 1:13
4
$\begingroup$

Where is the 1/2 come from?

This is because it is second order ode, and you have one B.C condition.

The general solution generates two constant of integrations.

So it is arbitrary which one of these two is solved for from the one condition given. DSolve choose to solve for C[1]. If you solve for C[2] instead, the 1/2 goes away.

ClearAll[y, x];
ode = 2 y'[x] + y''[x] == 0
bc = y[0] == 0
sol = y[x] /. First@DSolve[ode, y[x], x]

Mathematica graphics

Now apply the BC conditon on the solution

eq = 0 == sol /. x -> 0

Mathematica graphics

Compare the particular solution now when choosing to solve for either constant. Choosing to solve for C[1] gives

sol /. Solve[eq, C[1]]

Mathematica graphics

And Choosing to solve for C[2] gives

sol /. Solve[eq, C[2]]

Mathematica graphics

So both are valid solutions. But it depends on the choice used to solve for the constant of integration. DSolve choose to solve for C[1]. And since the constant that remains is arbitrary, it can be named anything else. say $C$. so you end up with any one of these two particular solutions

Mathematica graphics

Both are valid and equivalent solutions since $C$ is constant, it does not matter if one writes C or C/2 or C/10 or C/100. After using the second B.C., the same answer will result.

$\endgroup$

Not the answer you're looking for? Browse other questions tagged or ask your own question.