2
$\begingroup$

We can numerically integrate first order differential equations using Euler method like this: $$y_{n+1} = y_n + hf(t_n, y_n)$$

And with Implicit Euler like this: $$y_{n+1} = y_n + hf(t_{n+1},y _{n+1})$$

If I have a differential equation $y' - ky = 0$, I can integrate $y$ numerically using Implicit Euler: $$y_{n+1} = y_n + hky_{n+1}$$ $$y_{n+1} = y_n\frac{1}{1-hk}$$

But how I do use Implicit Euler for second order differential equations, like for instance the equation for simple harmonic motion? $$y'' + w^2y = 0$$

We have to integrate with respect to $y$ and $y'$. For explicit Euler the numerical integration would look like this (?): $$y_{n+1} = y_n + hf(t_n, y'_n)$$ $$y'_{n+1} = y'_n + hg(t_n, y_n)$$

How would we do integrate using Implicit Euler instead?

Improve this question
$\endgroup$
2
  • $\begingroup$ As you can see in the answer: always rewrite your system as a first order one. $\endgroup$ Commented Mar 10, 2020 at 13:53
  • $\begingroup$ Please avoid cross-postings, or add links to the other places on this network, like math.stackexchange.com/questions/3576061/… $\endgroup$ Commented Mar 10, 2020 at 15:52

1 Answer 1

Reset to default
6
$\begingroup$

You have to write your second order equation as a system of two first order equations. Let $y' = v$, then your equation

$$ y'' + \omega^2 y = 0 $$

becomes

$$ \begin{pmatrix} y' \\ v' \end{pmatrix} = \begin{pmatrix} v \\ -\omega^2 y \end{pmatrix} = \begin{pmatrix} 0 & 1 \\ -\omega^2 & 0 \end{pmatrix} \begin{pmatrix} y \\ v \end{pmatrix} $$

If you denote $u = (y,v)$ and the matrix as $A$, this system can be written as

$$ u' = A u $$

Now implicit Euler reads

$$ u_{n+1} = u_n + h A u_{n+1} $$

or

$$ \left( I - h A \right) u_{n+1} = u_n $$

Translated back into components you get

$$ \begin{pmatrix} 1 & -h \\ h \omega^2 & 1 \end{pmatrix} \begin{pmatrix} y_{n+1} \\ v_{n+1} \end{pmatrix} = \begin{pmatrix} y_{n} \\ v_n \end{pmatrix} $$

Addendum. It might be worth pointing out that implicit Euler is not a very good integrator for this type of problem as it will lead to artificial energy dissipation. You might be better of with what is called symplectic Euler method.

Improve this answer
$\endgroup$
1
  • $\begingroup$ Indeed, I failed to make an even number of sign errors :) Fixed. $\endgroup$ Commented Mar 11, 2020 at 8:11

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.