0
$\begingroup$

I try to solve this couple of second order differential equations to get an analytic solution for x(t) and y(t)

(x'[t]/x[t]) (y'[t]/3*y[t]) + y''[t]/3*y[t] + 2 x''[t]/x[t] =0

2*(x'[t]/x[t]) y'[t] + y''[t] = 0  

I'm a beginner in MA. However I think DSolve can not solve these coupled ODE in one step, here is my trial :

eq1[t_] := x'[t]/x[t] * y'[t]/(3 y[t]) + y''[t]/(3 y[t]) +  2 *x''[t]/x[t]
eq2[t_] := 2 x'[t] /x[t] y'[t] + y''[t]
DSolve[{eq1[t] == eq2[t] == 0, x[0] == 0, y[0] == 0}, {x, y}, t]

Which gives no solution. Noting that the IC are arbitrary. So any help how to go around that, I mean can the function x given in terms of y from eq1, then substitute in eq2 which will be in this case a function of y only.

$\endgroup$
1
  • $\begingroup$ Always try without IC first. If it can solve, then now add the IC's. No point of trying with IC if it can't solve it without IC. DSolve can't solve it without IC, so most likely this is unsolvable. May be you made mistake in entering the ode's from the text book ?. It is very easy to make up ode's which can not be solved analytically. $\endgroup$
    – Nasser
    Commented Oct 3 at 13:43

1 Answer 1

2
$\begingroup$

Your initial conditions: x[0]==0, y[0]==0 are not compatible with your equations that contain terms like: 1/x[t]and 1/y[t]. Further a second order DE needs 2 initial conditions for a specific solution, you need to add the initial values of the derivative.

You may try to solve these equations numerically by choosing suitable initial values. E.g.:

eq1[t_] := 
 x'[t]/x[t]*y'[t]/(3 y[t]) + y''[t]/(3 y[t]) + 2*x''[t]/x[t]
eq2[t_] := 2 x'[t]/x[t] y'[t] + y''[t]
sol[t_] = {x[t], y[t]} /. 
   NDSolve[{eq1[t] == 0, eq2[t] == 0, x[0] == 1, y[0] == 2, 
      x'[0] == 1, y'[0] == -1}, {x, y}, {t, 0, 5}][[1]];
Plot[sol[t], {t, 0, 10}, PlotRange -> {0, 5}]

enter image description here

$\endgroup$
1
  • $\begingroup$ Thanks so much for your answer @DanielHuber .. I was trying to extrapolate analytic solution from the numeric solution by FindFit or so $\endgroup$
    – Dr. phy
    Commented Oct 6 at 9:46

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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