3
$\begingroup$

So first I want to give you some background information:

begin of the background information

I'm currently reading an abstact about the Lotka Volterra differential equations:

$$ x^{'} = x -xy $$ $$ y^{'} = -y +xy $$

We know that the most numerical methods give us spiraling solutions instead of cyclic. So I want to try to show that a modified forward Euler Method leads to cyclic solutions.

Here:

$$ \frac{x_{n+1}-x_n}{\Delta t} = x_n -x_ny_n $$ $$\frac{y_{n+1}-y_n}{\Delta t} = -y_n -x_{n+1}y_n $$

There is a proof to show that this modification does not spiral.

end of background information

Now I have some questions:

So let us say to simplify notitation $$ X = \Delta tx +x - \Delta txy$$ and $$ Y = -\Delta ty+ y + \Delta tXy $$ where we set $X:=x_{n+1}, Y:=y_{n+1}$, $x:= x_n$ and $ y := y_n$ solved for the unknown $X$ and $Y$.

Taking derivatives: $$ dX = \Delta tdx +dx - \Delta tdxy - \Delta txdy $$ and $$ dY = -\Delta tdy+ dy + \Delta tdXy + \Delta tXdy $$

Finally we arrived to one point where I stucked.

I need that $ dX \wedge dX = 0 $ and $ dY \wedge dY = 0 $. Can you help me out here?

$\endgroup$
  • 1
    $\begingroup$ That is by the defining properties of an anti-symmetric product. What is your understanding of the wedge or outer product? $\endgroup$ – LutzL May 24 at 16:23
  • $\begingroup$ That sounds reasonable. Do you know what confuses me? In the abstract they say that this follows after some "manipulations". So I thought that I have to do something more. $\endgroup$ – RukiaKuchiki May 24 at 16:43
  • $\begingroup$ Usually, the symplectic Euler method only works for Hamiltonian systems $\dot x=\partial_yH(x,y)$, $\dot y=-\partial_xH(x,y)$. While the LV system has a first integral $F(x,y)=x-\ln x+y-\ln y$, it is not a Hamiltonian function of the system. $\endgroup$ – LutzL May 24 at 16:52
3
$\begingroup$

A defining property of the outer product is its anti-symmetry $$ a∧b = -b∧a, $$ which implies that $$ 2\,a∧a=0. $$

$\endgroup$
  • $\begingroup$ Dear LutzL, I found the property $ a \wedge b = (-1)^{kl} b \wedge a $. Sorry for the stupid question, but why do you have only $ -b \wedge a$ . Like I said I completely new at this topic. $\endgroup$ – RukiaKuchiki May 25 at 17:39
  • $\begingroup$ This elementary formula is for $a,b\in V=\Lambda^1V$. The extended formula you cite is for the case that $a,b$ are already higher order outer products, $a\in\Lambda^kV$ and $b\in\Lambda^lV$. $\endgroup$ – LutzL May 25 at 17:56

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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