I am doing this question
As it can't be solved using separation of variables (my assumption according to what i did, after checking by substituting $w(y,t)=f(y)g(t)$ , and getting a term at last which is not depending on only single variable)
Did my assumption is right ?
So how to solve this equation, I am stuck ?
To get rid of the non-homogeneous part, you can substitute another variable and another function.
First, we can substitute:
$$z=2y+1 $$
Which gives us:
$$4c w_{zz}-w_t+z=0$$
Then we introduce a new function:
$$w=v(z,t)-\frac{z^3}{24 c}$$
Substituting, we get:
$$4c v_{zz}-v_t=0$$
The boundary and initial conditions change in an obvious way, and the result is a separable equation for $v(z,t)$ which is easy enough to solve.
As a further hint, the conditions change to (if I didn't make a mistake somewhere):
$$v(0,t)=\frac{1}{24c}, \qquad v(2L+1,t)=\frac{(2L+1)^3}{24c}+4 \\ v(z,0)=\frac{z^3}{24c}+\Phi \left(\frac{z-1}{2} \right)$$