Skip to content

ichamail/2D_Subsonic_Potential_Aerodynamics

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Subsonic Potential Aerodynamics

A 2D Aerodynamic Potential-Flow Code

Theoretical Model

Potential Flow Theoretical Model

Velocity Field

  • incompressible: $\nabla \cdot \underline{V} = 0$
  • conservative: $\underline{V} = \nabla \phi \implies \nabla \times \underline{V} = \nabla \times \nabla \phi = 0$ (irrotational)
$$\nabla \cdot \underline{V} = \nabla \cdot \nabla \phi = \nabla^2 \phi = 0$$

Integral Equation of velocity potential $\phi$

$$\phi(\underline{r}_p) = \iint_S \frac{\sigma}{2\pi} \ln{(\lVert \underline{r} - \underline{r}_p \rVert)} \mathrm{d}S + \iint_{S \cup S_w} - \frac{\mu}{2\pi} (\underline{e}_n \cdot \nabla) \ln{(\lVert \underline{r} - \underline{r}_p \rVert)} \mathrm{d}S + \phi_\infty(\underline{r}_p)$$

where:

  • $\mu = \phi - \phi_i$
  • $\sigma = (\underline{e}_n \cdot \nabla)(\phi - \phi_i)$
  • $\phi_\infty(\underline{r}_p) = \iint_{S_\infty} \left[ (\underline{e}_n \cdot \nabla) \phi \ln{ \left( \frac{\lVert \underline{r} - \underline{r}_p \rVert}{2\pi} \right) } - \phi (\underline{e}_n \cdot \nabla) \ln{ \left( \frac{\lVert \underline{r} - \underline{r}_p \rVert}{2\pi} \right) } \right] \mathrm{d}S$

if $\mu = \phi - \phi_i = \phi - \phi_\infty$ and $\sigma = (\underline{e}_n \cdot \nabla)(\phi - \phi_i) = (\underline{e}_n \cdot \nabla)(\phi - \phi_\infty )$, then for $P \in S^-$ :

$$\iint_S \frac{\sigma}{2\pi} \ln{(\lVert \underline{r} - \underline{r}_p \rVert)} \mathrm{d}S + \iint_{S \cup S_w} - \frac{\mu}{2\pi} (\underline{e}_n \cdot \nabla) \ln{(\lVert \underline{r} - \underline{r}_p \rVert)} \mathrm{d}S = 0, \qquad \forall (x_p, y_p, z_p) \in S: (\underline{r} - \underline{r}_p) \cdot \underline{e}_n \to 0^+$$

note: $\sigma = (\underline{e}_n \cdot \nabla)(\phi - \phi_i) = (\underline{e}_n \cdot \nabla)(\phi - \phi_\infty ) = (\underline{e}_n \cdot \nabla)\phi - (\underline{e}_n \cdot \nabla)\phi_\infty = \underline{V} - \underline{e}_n \cdot \underline{V}_\infty = - \underline{e}_n \cdot \underline{V}_\infty$

Notable Observations:

  • A vector field $\underline{V}: U \to \mathbb{R}^n$, where $\mathbb{U}$ is an open subset of $\mathbb{R}^n$, is said to be conservative if there exists a $\mathrm{C}^1$ (continuously differentiable) scalar field $\phi$ on $\mathbb{U}$ such that $\underline{V} = \nabla \phi$.

  • According to Poincaré's Lemma, A continuously differentiable ($\mathrm{C}^1$) vector field $\underline{V}$ defined on a simply connected subset $\mathbb{U}$ of $\mathbb{R}^n$ ($\underline{V} \colon \mathbb{U} \subseteq \mathbb{R}^n \to \mathbb{R}^n$), is conservative if and only if it is irrotational throughout its domain ($\nabla \times \underline{V} = 0$, $\forall \underline{x} \in \mathbb{U}$).

  • Circulation $\Gamma = \oint_{C} \underline{V} \cdot \mathrm{d} \underline{l} = \iint_S \nabla \times \underline{V} \cdot \mathrm{d}\underline{S}$ In a conservative vector field this integral evaluates to zero for every closed curve. $\Gamma = \oint_{C} \underline{V} \cdot \mathrm{d} \underline{l} = \iint_S \nabla \times \underline{V} \cdot \mathrm{d}\underline{S} = \iint_S \nabla \times \nabla \phi \cdot \mathrm{d}\underline{S} = 0$

  • The space exterior to a 2D object (e.g. an airfoil) is not simply connected

Numerical Model (Panel Methods)

$$\sum_{j=0}^{N_s - 1} B_{ij} \sigma_j + \sum_{j=0}^{N_s + N_w - 1} C_{ij} \mu_j = 0 , \qquad 0 \le i < N_s$$

where:

  • $B_{ij} = \frac{1}{2\pi} \iint_{S_j} \ln{(\lVert \underline{r} - \underline{r}_{cp_i} \rVert)} \mathrm{d}{S_j} = \frac{1}{2\pi} \iint_{t_1}^{t_2} \ln \left(\sqrt{(t_{cp_i} - t)^2 + n_{cp_i}^2} \right) \mathrm{d}{t_j}$

  • $C_{ij} = \frac{1}{2\pi} \iint_{S_j} (\underline{e}_n \cdot \nabla) \ln{(\lVert \underline{r} - \underline{r}_{cp_i} \rVert)} \mathrm{d}{S_j} = \frac{1}{2\pi} \iint_{S_j} \frac{n_{cp_i}}{(t_{cp_i} - t)^2 + n_{cp_i}^2} \mathrm{d}{t_j}$

from Kutta Condition: $\mu_w = const = \mu_U - \mu_L$

$$A_{ij} \mu_j = - B_{ij} \sigma_j , \qquad A_{ij} = \begin{cases} C_{ij} + \sum\limits_{k} C_{ik} & \text{if $j=0$}\\\ C_{ij} & \text{if $0 < j < N_s - 1$}\\\ C_{ij} - \sum\limits_{k} C_{ik} & \text{if $j=N_s-1$} \end{cases}$$ $$0 \le i < N_s \qquad 0 \le j < N_s \qquad N_s \le k < N_s + N_w$$

Features

  1. Calculation of Non-Lifting Potential Flow about 2D arbitrarily-shaped rigid bodies

    1. Steady simulations
    2. Unsteady simulations
  2. Calculation of Lifting Pseudo-Potential Flow around 2D arbitrarily-shaped rigid bodies

    1. Steady state simulations with flat rigid wake model
    2. Steady state iterative simulations with flexible wake model
    3. Unsteady simulations with a shedding wake model

Simulation Results

Potential Flow around a 2D Circular Object

circle mesh in body-fixed frame

circle mesh in body-fixed frame

circle stream plot 10 panels

circle pressure coefficient 10 panel

circle pressure coefficient 20 panel

Potential Flow around an Airfoil

numerical model in inertial frame

numerical model in body-fixed frame

Steady Simulation with rigid wake

rigid wake

stream lines

pressure coefficient distribution

aerodynamic force vector

Steady Simulation with iterative wake

iterative wake

streamlines

pressure coefficient distribution

aerodynamic force vector

Unsteady Simulation with wake roll up

wake rollup

streamlines

wake rollup final step

pressure coefficient distribution

aerodynamic force vector

About

Panel Methods Aerodynamic Solver for 2D Potential Flows

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages