0
$\begingroup$

I am trying to solve a system of PDEs, but I get the error NDSolveValue::ntdvdae:

Cannot solve to find an explicit formula for the derivatives. NDSolve will try solving the system as differential-algebraic equations

After that Mathematica runs for a while and then spits out another error NDSolveValue::icfail:

Unable to find initial conditions that satisfy the residual function within specified tolerances. Try giving initial conditions for both values and derivatives of the functions.

Here is the code:

l0 = -1;
r0 = 1;
G = 6.67 10^-11;
c = 1;
c0 = 299792458;
P[t_, r_] := \[Epsilon][t, r]/3;
ic = {R[0, r] == r, Derivative[1, 0][R][0, r] == c, 
   Derivative[1, 0][\[Lambda]][0, r] == 
    1, \[Lambda][0, r] == (-(1/9) r + 10/9) l0, \[Nu][0, 
     r] == (-(1/9) r + 10/9) ((4 Pi)/3 10)^(-1/4), \[Epsilon][0, 
     r] == (4 Pi)/3 10};
bc = {R[t, r0] == r0 + c t, 
   R[t, 10] == 10 + c t, \[Lambda][t, r0] == l0, \[Lambda][t, 10] == 
    0, \[Nu][t, r0] == ((4 Pi)/3 10)^(-1/4), \[Nu][t, 10] == 
    0, \[Epsilon][t, r0] == (4 Pi)/3 10};
eq1 = D[\[Epsilon][t, r], t] R[t, 
      r] + (P[t, r] + \[Epsilon][t, r]) (2 D[R[t, r], t] + 
       R[t, r] D[\[Lambda][t, r], t]) == 0;
eq2 = (P[t, r] + \[Epsilon][t, r]) D[\[Nu][t, r], r] + 
    D[P[t, r], r] == 0;
eq3 = D[\[Nu][t, r], r] D[R[t, r], t] + 
    D[R[t, r], r] D[\[Lambda][t, r], t] - D[D[R[t, r], t], r] == 0;
eq5 = R[t, 
     r] (Exp[2 \[Nu][t, 
          r]] (-((D[\[Lambda][t, r] - \[Nu][t, r], 
              r]) (D[R[t, r], r] + R[t, r] D[\[Nu][t, r], r])) + 
         D[R[t, r], {r, 2}] + R[t, r] D[\[Nu][t, r], {r, 2}]) - 
      Exp[2 \[Lambda][t, 
          r]] (((D[\[Lambda][t, r] - \[Nu][t, r], t]) (D[R[t, r], t] +
              R[t, r] D[\[Lambda][t, r], t])) + D[R[t, r], {t, 2}] + 
         R[t, r] D[\[Lambda][t, r], {t, 2}])) == (8 \[Pi] G)/
    c0^4 Exp[2 \[Lambda][t, r] + 2 \[Nu][t, r]] P[t, r];
{en, Rad, lam, nu} = 
 NDSolveValue[{eq1, eq2, eq3,(*eq4,*)eq5, ic, bc}, {\[Epsilon], 
   R, \[Lambda], \[Nu]}, {t, 0, 10}, {r, r0, 10}]
$\endgroup$
1
  • $\begingroup$ Your pde-system is of order one concerning Derivative[1,0][f_][t,r]. Trying to find explicit form fails, system seems to be singular!. $\endgroup$ Commented Dec 23, 2023 at 18:39

1 Answer 1

1
$\begingroup$

eq2 can be solved symbolically.

seq2 = DSolve[{eq2, bc[[7]]}, ϵ[t, r], {t, r}][[1, 1]]
(* ϵ[t, r] -> 40/3 E^(4 γ[t, 1] - 4 γ[t, r]) Pi *)

Now, apply the ϵ[0, r] initial condition.

seq2 /. t -> 0 /. (ic[[6]] /. Equal -> Rule);
Solve[% /. Rule -> Equal, γ[0, r]][[1, 1]] /. C[1] -> 0
(* γ[0, r] -> γ[0, 1] *)

which is inconsistent with ic[[5]].

γ[0, r] == ((3/(5 Pi))^(1/4) (10/9 - r/9))/2^(3/4)

It is unlikely that the error messages reported in the question are caused by this inconsistency. Nonetheless, it is not possible to proceed until the inconsistency is resolved.

$\endgroup$

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.