
1) Fonctions polynomiales
2) Fonctions trigonométriques
3) Fonctions holomorphes et harmoniques
3.1) Fonctions holomorphes
Définition:Soient \(U\) un ouvert de l'ensemble \(\mathbb{C}\) des nombres complexes et \(f\) une application de \(U\) dans \(\mathbb{C}\).
• On dit que \(f\) est dérivable (au sens complexe) ou holomorphe en un point \(z_0\) de \(U\) si la limite suivante, appelée dérivée de \(f\) en \(z_0\) existe :
\( f'(z_0)= \displaystyle \lim_{z \to z_0} \dfrac{f(z)-f(z_0)}{z-z_0} \).
• On dit que \(f\) est holomorphe sur \(U\) si elle est holomorphe en tout point de \(U\).
• En particulier, on appelle fonction entière une fonction holomorphe sur \(\mathbb{C}\).
Conditions de Cauchy-Riemann
Soit \(f(z)=P(x,y)+iQ(x,y)\).Si \( \frac{\partial P}{\partial x}(x,y)\), \( \frac{\partial P}{\partial y}(x,y) \), \( \frac{\partial Q}{\partial x}(x,y)\), \(\frac{\partial Q}{\partial y}(x,y)\) existent et sont continues.
Et si:
\( \left\{ \begin{array}{ll} \displaystyle \frac{\partial P}{\partial x}=\frac{\partial Q}{\partial y} \\ \displaystyle \frac{\partial P}{\partial y}=-\frac{\partial Q}{\partial x} \end{array} \right. \)
Alors, \(f\) est holomorphe sur \(\mathbb{C}\), et \(\displaystyle f'(z)=\frac{\partial f}{\partial x}=\frac{\partial P}{\partial x}(x,y) + i \frac{\partial Q}{\partial x}(x,y) \).
La dérivée d'une fonction complexe est donnée par \(f'(z)= \displaystyle \lim_{h \to 0} \dfrac{f(z+h) - f(z)}{h} \).
Notons \(h=s+iv\) cet accroissement. Il peut se calculer le long de \(s\) et le long de \(iv\).
Le long de \(s\), cet accroissement vaut:
\(f'(z)= \displaystyle \lim_{s \to 0} \dfrac{f(z+s) - f(z)}{s} = \frac{\partial f}{\partial x} = \frac{\partial P}{\partial x} + i \frac{\partial Q}{\partial x} \)
Le long de \(iv\), cet accroissement vaut:
\(f'(z)= \displaystyle \lim_{v \to 0} \dfrac{f(z+iv) - f(z)}{iv} = -i \frac{\partial f}{\partial y} = \frac{\partial Q}{\partial y} - i \frac{\partial P}{\partial y} \)
L'unicité de cette dérivée implique l'égalité:
\(\displaystyle \frac{\partial f}{\partial x} = -i \frac{\partial f}{\partial y} \) qui peut encore s'écrire \(\displaystyle \frac{\partial P}{\partial x} = \frac{\partial Q}{\partial y} \) et \(\displaystyle \frac{\partial P}{\partial y} = - \frac{\partial Q}{\partial x} \).
3.2) Fonctions harmoniques
Définition:Soit \(\varphi : \mathbb{R^2} \rightarrow \mathbb{R}\)
\( (x,y) \mapsto \varphi (x,y) \) où \( (x,y) \mapsto \varphi (x,y) \) est de classe \(\mathcal C^2\), autrement dit les fonctions \( \frac{\partial^2 \varphi}{\partial x^2} \), \( \frac{\partial^2 \varphi}{\partial y^2} \), \( \frac{\partial^2 \varphi}{\partial x \partial y} \) existent et sont continues.
Lorsqu'on a ces conditions, le théorème de Schwartz s'applique.
Théorème:
Soit U un ouvert de \(\mathbb{R^n}\) et \(f\) une fonction de U dans \(\mathbb{R^p}\) de classe \(\mathcal C^2\) sur U. Alors, pour tout \(a\) dans U et tous \(i,j\) dans {1,...,n} on a:
\( \frac{\partial^2 f}{\partial x_i \partial x_j} (a) = \frac{\partial^2 f}{\partial x_j \partial x_i} (a) \).
Dans notre cas, cela signifie \( \displaystyle \frac{\partial^2 \varphi}{\partial x \partial y} = \frac{\partial^2 \varphi}{\partial y \partial x} \)
\(\varphi\) est dite harmonique \(\displaystyle \iff \frac{\partial^2 \varphi}{\partial x^2} + \frac{\partial^2 \varphi}{\partial y^2} = 0\), c'est-à-dire si seulement si le laplacien de \(\varphi\) est nul \( \Delta \varphi = 0 \).
Théorème:
Soit \( f(z)=P(x,y)+iQ(x,y) \) une fonction holomorphe.
Alors \(P(x,y)\) et \(Q(x,y)\) sont harmoniques.
Supposons \( f(z) \) holomorphe, donc:
\( \frac{\partial P}{\partial x} = \frac{\partial Q}{\partial y}\) et \( \frac{\partial P}{\partial y} = - \frac{\partial Q}{\partial x}\).
\( \frac{\partial^2 P}{\partial x^2} = \frac{\partial }{\partial x}(\frac{\partial Q}{\partial y}) = \frac{\partial^2 Q}{\partial x \partial y}\)
\( \frac{\partial^2 P}{\partial y^2} = \frac{\partial }{\partial y}(-\frac{\partial Q}{\partial x}) = - \frac{\partial^2 Q}{\partial y \partial x} = - \frac{\partial^2 Q}{\partial x \partial y} \)
\( \Delta P = \frac{\partial^2 P}{\partial x^2} + \frac{\partial^2 P}{\partial y^2} = 0\), donc \(P\) est harmonique.
Théorème:
Soit \( P(x,y) \) harmonique
Alors il existe une fonction \( f : \mathbb{C} \rightarrow \mathbb{C}\) holomorphe telle que \( \Re (z)=P(x,y) \) ou \( \Im (z)=P(x,y) \).
Exercices
4) Fonction W de Lambert
La fonction \(W\) de Lambert parfois aussi appelée la fonction Oméga, est la fonction définie sur \([-\frac{1}{e};+\infty[\) par:
\(W(xe^x)=x\)
La fonction \(W\) de Lambert ne peut pas être exprimée à l'aide de fonctions élémentaires.
Les valeurs remarquables sont :
➨\( W(0)= 0\)
➨\( W(-e^{-1})= -1\)
➨\( W(1)= \Omega\) où \( \Omega \) est le réel positif défini par \( \Omega e^{\Omega}=1 \)
\(\Omega \approx 0,5671432904\)
Beaucoup d'équations impliquant des exponentielles peuvent être résolues par l'utilisation de la fonction \(W\). La stratégie générale est de déplacer toutes les instances de l'inconnue d'un côté de l'équation et de le faire ressembler à \(xe^x\). À ce stade, la « fonction » \(W\) nous fournit les solutions : \(X e^X = Y ⟺ X = W ( Y )\)
(chaque branche différente de la « fonction » \(W\) donne une solution différente).
1)