You can use separation of variables, but you may need the second order derivative for a boundary condition.
– Chinny84Mar 03 '14 at 07:59
1
Is $L_2(t)$ specified? That's the sticking point: BTW it makes separation of variables impossible.
– Robert IsraelMar 03 '14 at 16:02
I have the feeling that the corresponding elliptic equation has no solution in $L^2(0L)$ (let alone on an interval with moving boundary), i.e., the spectrum of the differential operator is too large for the third derivative to generate a semigroup on $L^2(0,L)$; and therefore there is no solution for general $F_0$.
– DeMJan 13 '16 at 15:56