In my Fluid Mechanics class, we arrived at an equation for 2D compressible flow (using perturbation theory):
$$\left(1-M^{2}\right)\frac{\partial^{2}\phi}{\partial x^{2}}+\frac{\partial^{2}\phi}{\partial y^{2}}=0$$
where $\phi$ is the perturbed fluid potential ($\vec v=\vec U+\vec\nabla(\phi)$), $\vec U=U\vec e_{x}$ is a constant flow and $M=U/c_s$ is the Mach number.
I wanted to resolve this equation for an arbitrary contour defined by $$ f_\pm=\begin{cases}f(x) & \quad\text{if }x\in[0,l] \\ 0 & \quad \text{if } x\in]l,+\infty[\end{cases} $$
where $f_{+}$ is for $y>0$ and $f_{-}$ for $y<0$. and you also asume $\frac{df}{dx}\ll(\text{the other non pertubed quantities})$
My teacher only solved the supersonic case using the fact that for $M>1$ you get the wave equation. He found his boundary conditions like this:
$$\frac{df}{dx}=\tan(\alpha)=\frac{\nabla\phi|_{y}}{U+\nabla\phi|_{y}}$$
wich gives neglecting the quadratic perturbed quantities:
$$\frac{\partial\phi}{\partial y}|_{y=f'(x)}\simeq\frac{\partial\phi}{\partial y}_{y=0}\simeq Uf'(x).$$
My problem is that I have tried solving this equation (using Fourier transform) for a subsonic flow using the same boundary condition (and a bounded solution). And while I found a solution, this solution doesn't seem very physical or accurate when I plot it on python.
So I was wondering if my boundary conditions are incorrect and if they are what are the correct one?
