« Utilisateur:Ellande/Brouillon3 » : différence entre les versions

Contenu supprimé Contenu ajouté
m Robot : Remplacement de texte automatisé (-\n(==={0,3})(?: *)([^\n=]+)(?: *)\1(?: *)\n +\n\1 \2 \1\n)
Aucun résumé des modifications
Balise : Éditeur de wikicode 2017
 
Ligne 1 :
{{Utilisateur:Ellande/Carnet brouillon}}
 
Conservation de la quantité de mouvement (équation d'Euler) :
==  Expression analytique pour un objectif parfait ==
L'expression analytique de la fonction de transfert d'un objectif possédant une pupille '''circulaire'''.
 
<math>\rho_0\cdot \frac {\mathrm d \vec v}{\mathrm d t} = -\overrightarrow \nabla p</math>.
'''[[Fréquence de coupure]] dans l'espace Image'''
: <math>f_c = \frac{1}{\lambda\cdot N} </math>
avec
 
Hypothèse simplificatrice : dans le cadre de l'hypothèse des petits mouvements, on néglige les termes du second ordre <math>(\vec v \cdot \overrightarrow \nabla)\vec v
<math>f_c</math> : fréquence de coupure optique dans l'espace image (détecteur)
</math> représentant l'accélération convective, due à la variation spatiale de la vitesse (dans son mouvement la particule passe par des points de vitesses différentes) : on suppose que l'accélération convective est négligeable face à l'accélération locale.
 
<math>\frac {\mathrm d \vec v}{\mathrm d t}
<math>\lambda</math> : longueur d'onde (on utilise couramment 550 nm pour le visible)
= \frac {\partial \vec v}{\partial t}
+ (\vec v \cdot \overrightarrow \nabla)\vec v
\simeq \frac {\partial \vec v}{\partial t} </math>.
 
Compte tenu des hypothèses, l'équation d'Euler peut alors s'écrire :
<math>N</math> : nombre d'ouverture de l'objectif ([[Ouverture (photographie)|F-number]])
 
<math>\rho_0\cdot \frac {\partial \vec v}{\partial t} = -\overrightarrow \nabla p</math>.
On note :<math>f_n = \frac{f}{f_c}</math> : la fréquence normalisée
 
== Onde sphérique ==
<math>\mathbf{FTM}(f_n) = \frac{2}{\pi} \left( \arccos(f_n) - f_n \sqrt{1-f_n^2} \right)</math>
Hypothèse : l'onde de pression est longitudinale : <math>\frac{\partial p}{\partial \theta}=0
</math> et <math>\frac{\partial p}{\partial \phi}=0
</math>.
 
<math>\overrightarrow{\mathrm{grad}}\ p =
== Critère de Rayleigh ==
\overrightarrow{\nabla} p =
<math>f_{lim} = \frac{1}{1,22\cdot \lambda\cdot N} </math>
\frac{\partial p}{\partial r}\overrightarrow{e_r} +
\frac{1}{r}\frac{\partial p}{\partial \theta}\overrightarrow{e_\theta} +
\frac{1}{r \sin\theta}\frac{\partial p}{\partial \phi}\overrightarrow{e_\phi}
=\frac{\partial p}{\partial r}\overrightarrow{e_r}
</math>
 
Hypothèse : l'onde de vitesse est longitudinale : <math>\vec v = v \ \vec e_r
Le MTF vaut 9 % à la limite de résolution de Rayleigh.
</math>.
 
L'équation d'Euler peut alors s'écrire : <math>\rho_0 \frac {\partial v}{\partial t} = -\frac {\partial p}{\partial r}</math>.
== Produit de convolution ==
 
=== Onde progressive sinusoïdale ===
{{Loupe|marge=0em|largeur=auto|icône=OOjs UI icon logo-wikipedia.svg|contenu=Plus de détails dans l’article wikipédia : '''[[w:fr:Produit de convolution|Produit de convolution]]'''.}}
Hypothèse : l'onde de pression s'atténue en 1/r.
 
<math>p(r,t)=\frac{\hat P(1\,\mathrm m)}{r}\cdot \cos (\omega\,t-k\,r+\varphi)
:<math> (f\ast g) (t) = \int_{-\infty}^{+\infty} f(t-\tau) \cdot g(\tau) \mathrm{d}\tau = \int_{-\infty}^{+\infty} f(\tau) \cdot g(t-\tau) \mathrm{d}\tau = (g\ast f) (t)</math>
=P(r)\sqrt 2\cdot \cos (\omega\,t-k\,r+\varphi)</math>
 
Sous forme complexe pour simplifier les calculs.
Pour une fonction périodique de période <math>T</math>
 
<math>\underline p(r,t)=P(r)\cdot \mathrm e^{\mathrm j(\omega\,t-k\,r+\varphi)}
:<math> (f\ast g) (t) = \frac 1 {T} \cdot \int_{t_0}^{t_0+T} f(t-\tau) \cdot g(\tau) \mathrm{d}\tau = \frac 1 {T} \cdot \int_{t_0}^{t_0+T} f(\tau) \cdot g(t-\tau) \mathrm{d}\tau</math>
=\frac{ P(1\,\mathrm m)}{r}\cdot \mathrm e^{\mathrm j(\omega\,t-k\,r+\varphi)}
</math>
 
==== Expression de la vitesse acoustique ====
Par dérivation
 
<math> \frac {\partial p}{\partial r}
* [[Transformation de Fourier]]
=-\frac {P(1\,\mathrm {m} )}{r^2}\cdot \cos (\omega\,t-k\,r+\varphi)
* [[Égalité de Parseval]]
+k\cdot\frac {P(1\,\mathrm {m} )}{r}\cdot \sin (\omega\,t-k\,r+\varphi)
* [[Théorème de Plancherel]]
 
* [[Produit de convolution]]
 
* {{Google Livres|mChYZgvldSgC|titre=Traitement des signaux et acquisition de données|page = 54}}
</math>
 
<math> \frac {\partial v}{\partial t}
=-\frac 1 {\rho_0} \frac {\partial p}{\partial r}
=-\frac 1 {\rho_0} \cdot\frac {P(1\,\mathrm {m} )}{r}\left(
-\frac {1}{r}\cdot \cos (\omega\,t-k\,r+\varphi)
+k\cdot \sin (\omega\,t-k\,r+\varphi)
\right)
</math>
 
En cherchant la primitive :
 
<math> v=\frac 1 {\rho_0} \cdot\frac {P(1\,\mathrm {m} )}{r}\left(
\frac {1}{\omega\, r}\cdot \sin (\omega\,t-k\,r+\varphi)
+\frac {k}{\omega}\cdot \cos (\omega\,t-k\,r+\varphi)
\right)
</math>
 
<math> v=\frac 1 {\rho_0\, c} \cdot\frac {P(1\,\mathrm {m} )}{r}\left(
\frac {1}{k\, r}\cdot \sin (\omega\,t-k\,r+\varphi)
+ \cos (\omega\,t-k\,r+\varphi)
\right)
</math>
 
<math> v=\frac 1 {\rho_0\, c} \cdot\frac {\sqrt{1+k^2\,r^2}}{k\, r}\cdot\frac {P(1\,\mathrm {m} )}{r}\left(
\frac {1}{\sqrt{1+k^2\,r^2}}\cdot \sin (\omega\,t-k\,r+\varphi)
+ \frac {k\, r}{\sqrt{1+k^2\,r^2}} \cos (\omega\,t-k\,r+\varphi)
\right)
</math>
 
<math> v=\frac 1 {\rho_0\, c} \cdot\frac {\sqrt{1+k^2\,r^2}}{k\, r}\cdot\frac {P(1\,\mathrm {m} )}{r}\left(
\sin \psi\cdot \sin (\omega\,t-k\,r+\varphi)
+ \cos\psi\cdot \cos (\omega\,t-k\,r+\varphi)
\right)
</math>
 
<math> v=\frac 1 {\rho_0\, c} \cdot\frac {\sqrt{1+k^2\,r^2}}{k\, r}\cdot\frac {P(1\,\mathrm {m} )}{r}\cdot \cos (\omega\,t-k\,r+\varphi-\psi)
 
</math>
 
==== Expression de la vitesse acoustique complexe ====
<math> \frac {\partial \underline p}{\partial r}
=\left(
-\frac {P(1\,\mathrm {m} )}{r^2}\cdot \mathrm {e} ^{-\mathrm {j} k\,r}
-\mathrm j\,k\cdot\frac {P(1\,\mathrm {m} )}{r}\cdot \mathrm {e} ^{-\mathrm {j} k\,r}
\right)
\cdot \mathrm {e} ^{\mathrm {j} (\omega \,t +\varphi) }
</math>
 
<math> \frac {\partial \underline v}{\partial t}
=-\frac 1 {\rho_0} \frac {\partial \underline p}{\partial r}
=\frac 1 {\rho_0} \left(
\frac {1}{r}+\mathrm j\,k
\right)
\cdot \frac {P(1\,\mathrm {m})}{r}\cdot\mathrm {e}^{\mathrm {j} (\omega \,t -k\,r+\varphi) }
</math>
 
<math> \underline v
=\frac 1 {\mathrm j \,\omega \,\rho_0} \left(
\frac {1}{r}+\mathrm j\,k
\right)
\cdot \frac {P(1\,\mathrm {m})}{r}\cdot\mathrm {e}^{\mathrm {j} (\omega \,t -k\,r+\varphi) }
</math>
 
Sachant que <math>
\frac {k}{\omega}=\frac {1}{c}
</math> alors :
 
<math> \underline v
=\frac 1 {\mathrm j \,\rho_0\,c} \left(
\frac {1}{k\,r}+\mathrm j
\right)
\cdot \underline p
</math>
 
==== Expression de l'impédance acoustique complexe ====
<math> \underline Z = \frac{\underline p}{\underline v}=
\mathrm {j} \,\rho _{0}\,c\left({\frac {k\,r}{1+\mathrm {j}\,k\,r}}\right)
 
 
</math>
 
<math> Z = |\underline Z| =\rho _{0}\,c\left({\frac {k\,r}{\sqrt{1+k^2\,r^2}}}\right)
 
 
</math>
 
<math> \psi=\mathrm{Arg}( \underline Z)
 
 
</math>
 
<math> \cos \psi ={\frac {k\,r}{\sqrt{1+k^2\,r^2}}}
 
 
</math>
 
==== Expression de l'intensité acoustique ====
<math> \vec I (r) = \frac 1 T \int_0^T p(r,t)\, \vec v(r,t)\,\mathrm d t
</math>
 
<math>
p\,v={\frac {1}{\rho _{0}\,c}}\cdot {\frac {\sqrt {1+k^{2}\,r^{2}}}{k\,r}}\cdot\frac {P^2(1\,\mathrm {m} )}{r^2}
\cos (\omega\,t-k\,r+\varphi)\cos (\omega\,t-k\,r+\varphi-\psi)\,\mathrm d t
 
 
</math>
 
<math>
p\,v={\frac {1}{\rho _{0}\,c}}\cdot {\frac {\sqrt {1+k^{2}\,r^{2}}}{k\,r}}\cdot\frac {P^2(1\,\mathrm {m} )}{r^2}
\left( \cos (-\psi)+\cos (2\,\omega\,t-2\,k\,r+2\,\varphi-\psi)\right)
 
 
 
</math>
 
<math>
I(r)={\frac {1}{\rho _{0}\,c}}\cdot {\frac {\sqrt {1+k^{2}\,r^{2}}}{k\,r}}
\cdot\frac {P^2(1\,\mathrm {m} )}{r^2} \cos \psi
={\frac {P^2(r)}{\rho _{0}\,c}}
 
 
 
</math>
 
==== Expression de l'intensité acoustique complexe ====
<math> \underline I (r) = \int_0^T \underline p(r,t)\, \underline v^*(r,t)\,\mathrm d t
</math>
 
<math> \underline I(r)
=\frac 1 {-\mathrm j \,\rho_0\,c }
\frac {1-\mathrm j\,k\,r}{k\,r}
\cdot\underline p^*\cdot\underline p
=\frac 1 {\rho_0\,c }
\frac {\mathrm j+k\,r}{k\,r}
\cdot P^2(r)
</math>
 
<math> I(r)=\mathrm{Re}(\underline I(r))
=\frac {P^2(r)} {\rho_0\,c }
</math>
 
== À voir / À creuser ==
Équation de continuité : <math>\frac{\partial \rho}{\partial t}+\hbox{div}(\rho \, \vec {v})=0</math>
 
Hypothèse : l'onde de vitesse est longitudinale. <math>\frac{\partial p}{\partial \theta}=0
</math> et <math>\frac{\partial p}{\partial \phi}=0
</math>.
 
<math>\mathrm{div}\, (\rho \vec v) =
\overrightarrow{\nabla} \cdot (\rho \vec v) =
\frac 1 {r^2} \frac{\partial}{\partial r}(r^2 \rho v_r) +
\frac1{r\sin\theta}\frac{\partial \sin\theta \rho v_\theta}{\partial\theta} +
\frac1{r\sin\theta}\frac{\partial \rho v_\phi}{\partial\phi}
=\frac 1 {r^2} \frac{\partial}{\partial r}(r^2 \rho v)</math>
 
Équation d'onde : <math>\Delta p - \frac 1 c \frac {\partial^2p}{\partial t ^2} = 0</math>.
 
Hypothèse : l'onde de pression est longitudinale. <math>\frac{\partial p}{\partial \theta}=0
</math> et <math>\frac{\partial p}{\partial \phi}=0
</math>.
 
<math>\Delta p =
\overrightarrow{\nabla}^2 M =
\frac{1}{r^2} \frac{\partial}{\partial r}\left( r^2 \frac{\partial p}{\partial r} \right) +
\frac{1}{r^2 \sin \theta} \frac{\partial}{\partial \theta} \left( \sin \theta \frac{\partial p}{\partial \theta} \right) +
\frac{1}{r^2 \sin^2 \theta} \frac{\partial^2 p}{\partial \phi^2}
=\frac{1}{r^2} \frac{\partial}{\partial r}\left( r^2 \frac{\partial p}{\partial r} \right)
</math>