Formelsammlung Tensoralgebra

Formelsammlung Tensoralgebra

$ {\sqrt[{n}]{x}} $ Dieser Artikel ist eine Formelsammlung zum Thema Tensoralgebra. Es werden mathematische Symbole verwendet, die im Artikel Liste mathematischer Symbole erläutert werden.

Diese Formelsammlung fasst Formeln und Definitionen der Tensoralgebra für Tensoren zweiter Stufe in der Kontinuumsmechanik zusammen. Es wird der dreidimensionale Raum zugrunde gelegt.

Allgemeines

Notation

  • Operatoren wie $ \mathrm {I} _{1} $ werden nicht kursiv geschrieben.
  • Buchstaben die als Indizes benutzt werden:
    • $ i,j,k,l,m,n\in \{1,2,3\} $.
      Ausnahme:
      Die imaginäre Einheit $ \mathrm {i} ^{2}=-1 $ und die #Vektorinvariante $ {\vec {\mathrm {i} }} $ werden in Abgrenzung zu den Indizes nicht kursiv geschrieben.
    • $ p,q,r,s\in \{1,2,\ldots ,9\} $
    • $ u,v\in \{1,2,\ldots ,6\} $
  • Alle anderen Buchstaben stehen für reelle Zahlen oder komplexe Zahlen.
  • Vektoren:
    • Alle hier verwendeten Vektoren sind geometrische Vektoren im dreidimensionalen euklidischen Vektorraum $ \mathbb {V} $.
    • Vektoren werden mit Kleinbuchstaben bezeichnet.
      Ausnahme #Dualer axialer Vektor $ {\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
    • Einheitsvektoren mit Länge eins werden wie in ê mit einem Hut versehen. Die Standardbasis von $ \mathbb {V} $ ist ê1,2,3.
    • Vektoren mit unbestimmter Länge werden wie in $ {\vec {a}} $ mit einem Pfeil versehen.
    • Dreiergruppen von Vektoren wie in $ {\vec {h}}_{1},{\vec {h}}_{2},{\vec {h}}_{3} $ oder $ {\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3} $ bezeichnen eine rechtshändige Basis von $ \mathbb {V} $.
    • Gleichnamige Basisvektoren mit unterem und oberem Index sind dual zueinander, z. B. $ {\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3} $ ist dual zu $ {\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3} $.
  • Tensoren zweiter Stufe werden wie in A mit fetten Großbuchstaben notiert. Die Menge aller Tensoren wird mit $ {\mathcal {L}}:=\mathrm {Lin} (\mathbb {V} ,\mathbb {V} ) $ bezeichnet. Tensoren höherer Stufe werden mit einer hochgestellten Zahl wie in $ {\stackrel {4}{\mathbf {C} }} $ geschrieben. Tensoren vierter Stufe sind Elemente der Menge $ {\stackrel {4}{\mathcal {L}}}:=\mathrm {Lin} ({\mathcal {L}},{\mathcal {L}}) $.
  • Es gilt die Einstein'sche Summenkonvention ohne Beachtung der Indexstellung.
    • Kommt in einer Formel in einem Produkt ein Index doppelt vor wie in $ c=a_{i}b^{i} $ wird über diesen Index summiert:
      $ c=a_{i}b^{i}=\sum _{i=1}^{3}a_{i}b^{i} $.
    • Kommen mehrere Indizes doppelt vor wie in $ c=A_{pq}B_{q}^{p} $ wird über diese summiert:
      $ c=A_{pq}B_{q}^{p}=\sum _{p=1}^{9}\sum _{q=1}^{9}A_{pq}B_{q}^{p} $.
    • Ein Index, der nur einfach vorkommt wie $ u $ in $ a_{u}=A_{uv}b_{v} $, ist ein freier Index. Die Formel gilt dann für alle Werte der freien Indizes:
      $ a_{u}=A_{uv}b_{v}\quad \leftrightarrow \quad a_{u}=\sum _{v=1}^{6}A_{uv}b_{v}\quad \forall \;u\in \{1,\ldots ,6\} $.

Glossar

Reservierte und besondere Symbole

Formelzeichen Abschnitt in der Formelsammlung Wikipedia-Artikel
$ \mathbf {I,1} $ #Einheitstensor Einheitstensor
$ \mathbf {Q,R} $ #Orthogonale Tensoren Orthogonaler Tensor
$ \lambda $ #Eigenwerte Eigenwertproblem
$ \delta _{ij} $ #Kronecker-Delta Kronecker-Delta
$ \epsilon _{ijk} $ #Permutationssymbol Permutationssymbol
$ {\stackrel {3}{\mathbf {E} }} $ #Fundamentaltensor 3. Stufe Epsilon-Tensor
$ [{\vec {a}}]_{\times } $ #Axialer Tensor oder Kreuzproduktmatrix Kreuzprodukt
$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}},\mathbf {A} _{\times } $ #Dualer axialer Vektor Kreuzprodukt
$ {\vec {\mathrm {i} }} $ #Vektorinvariante Vektorinvariante
$ \mathrm {i} $ Imaginäre Einheit

Zeichen für Operatoren

Formelzeichen Abschnitt in der Formelsammlung Wikipedia-Artikel
$ (\cdot )\cdot (\cdot ) $ Skalarprodukt von Vektoren, #Vektortransformation, #Tensorprodukt Skalarprodukt
$ (\cdot )\times (\cdot ) $ #Kreuzprodukt eines Vektors mit einem Tensor, #Kreuzprodukt von Tensoren Kreuzprodukt
$ (\cdot ):(\cdot ) $ #Skalarprodukt von Tensoren Frobenius-Skalarprodukt
$ (\cdot )\otimes (\cdot ) $ #Dyadisches Produkt Dyadisches Produkt
$ (\cdot )\cdot \!\!\times (\cdot ) $ #Skalarkreuzprodukt von Tensoren
$ (\cdot )\times \!\!\times (\cdot ) $ #Doppeltes Kreuzprodukt von Tensoren
$ (\cdot )\#(\cdot ) $ #Äußeres Tensorprodukt Äußeres Tensorprodukt
$ \parallel (\cdot )\parallel $ #Betrag Frobeniusnorm
$ |x|,|{\vec {v}}|,|\mathbf {A} | $ Betrag der Zahl x oder des Vektors $ {\vec {v}} $, #Determinante des Tensors A Determinante

Tensorfunktionen

Formelzeichen Abschnitt in der Formelsammlung Wikipedia-Artikel
$ \mathrm {Sp,tr,I} _{1} $ #Spur Spur (Mathematik), Hauptinvariante
$ \mathrm {I} _{2} $ #Zweite Hauptinvariante Hauptinvariante
$ \mathrm {det,I} _{3},|\mathbf {A} | $ #Determinante Determinante, Hauptinvariante
sym #Symmetrischer Anteil Symmetrische Matrix
skw, skew #Schiefsymmetrischer Anteil Schiefsymmetrische Matrix
adj #Adjunkte Adjunkte
cof #Kofaktor Minor (Mathematik)#Kofaktormatrix
dev #Deviator Deviator, Spannungsdeviator
sph #Kugelanteil Kugeltensor

Indizes

Formelzeichen Abschnitt in der Formelsammlung Wikipedia-Artikel
$ (\cdot )_{ij},(\cdot )^{ij},(\cdot )_{j}^{i} $ #Tensorkomponenten
$ (\cdot )^{\top } $ #Transposition Transponierte Matrix
$ (\cdot )^{\stackrel {mn}{\top }} $ Transpositionen von Tensoren vierter Stufe
$ (\cdot )^{-1} $ #Inverse Inverse Matrix
$ (\cdot )^{-\top },(\cdot )^{\top -1} $ #Transposition der #Inverse
$ (\cdot )^{\mathrm {S} } $ #Symmetrischer Anteil Symmetrische Matrix
$ (\cdot )^{\mathrm {A} } $ #Schiefsymmetrischer Anteil Schiefsymmetrische Matrix
$ (\cdot )^{\mathrm {D} } $ #Deviator Deviator, Spannungsdeviator
$ (\cdot )^{\mathrm {K} } $ #Kugelanteil Kugeltensor
$ {\stackrel {n}{(\cdot )}} $ Tensor n-ter Stufe
$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}},\mathbf {A} _{\times } $ #Dualer axialer Vektor Kreuzprodukt

Mengen

Formelzeichen Elemente
$ \mathbb {R} $ Reelle Zahlen
$ \mathbb {C} $ Komplexe Zahlen
$ \mathbb {V} $ Vektoren
$ {\mathcal {L}}=\mathrm {Lin} (\mathbb {V,V} ) $ Tensoren zweiter Stufe
$ {\stackrel {4}{\mathcal {L}}}=\mathrm {Lin} ({\mathcal {L,L}}) $ #Tensoren vierter Stufe

Kronecker-Delta

$ \delta _{ij}=\delta ^{ij}=\delta _{i}^{j}=\delta _{j}^{i}={\begin{cases}1&\mathrm {falls} \quad i=j\\0&\mathrm {sonst} \end{cases}} $

Für Summen gilt dann z. B.

$ v_{i}\delta _{ij}=v_{j} $
$ A_{ij}\delta _{ij}=A_{ii} $

Dies gilt für die anderen Indexgruppen entsprechend.

Permutationssymbol

$ \epsilon _{ijk}={\hat {e}}_{i}\cdot ({\hat {e}}_{j}\times {\hat {e}}_{k})={\begin{cases}1&{\text{falls}}\;(i,j,k)\in \{(1,2,3),(2,3,1),(3,1,2)\}\\-1&{\text{falls}}\;(i,j,k)\in \{(1,3,2),(2,1,3),(3,2,1)\}\\0&{\text{sonst, d.h. bei doppeltem Index}}\end{cases}} $
$ \epsilon _{ijk}\epsilon _{lmn}={\begin{vmatrix}\delta _{il}&\delta _{jl}&\delta _{kl}\\\delta _{im}&\delta _{jm}&\delta _{km}\\\delta _{in}&\delta _{jn}&\delta _{kn}\end{vmatrix}} $
$ \epsilon _{ijk}\epsilon _{klm}=\delta _{il}\delta _{jm}-\delta _{im}\delta _{jl} $
$ \epsilon _{ijk}\epsilon _{jkl}=2\delta _{il} $
$ \epsilon _{ijk}\epsilon _{ijk}=6 $

Kreuzprodukt:

$ a_{i}{\hat {e}}_{i}\times b_{j}{\hat {e}}_{j}=\epsilon _{ijk}a_{i}b_{j}{\hat {e}}_{k}=\epsilon _{ijk}a_{j}b_{k}{\hat {e}}_{i}=\epsilon _{ijk}a_{k}b_{i}{\hat {e}}_{j} $
$ \epsilon _{ijk}{\hat {e}}_{k}={\hat {e}}_{i}\times {\hat {e}}_{j} $

Spaltenvektoren und Matrizen

Die hier verwendeten Vektoren sind Spaltenvektoren

$ {\vec {a}}=a_{i}{\hat {e}}_{i}={\begin{pmatrix}a_{1}\\a_{2}\\a_{3}\end{pmatrix}} $

Drei Vektoren $ {\vec {a}},{\vec {b}},{\vec {c}} $ können spaltenweise in einer 3×3-Matrix $ M $ arrangiert werden:

$ M={\begin{pmatrix}{\vec {a}}&{\vec {b}}&{\vec {c}}\end{pmatrix}}={\begin{pmatrix}a_{1}&b_{1}&c_{1}\\a_{2}&b_{2}&c_{2}\\a_{3}&b_{3}&c_{3}\end{pmatrix}} $

Die Determinante der Matrix

$ |M|={\begin{vmatrix}{\vec {a}}&{\vec {b}}&{\vec {c}}\end{vmatrix}} $

ist

  • ungleich null, wenn die Spaltenvektoren linear unabhängig sind und
  • größer null, wenn die Spaltenvektoren zusätzlich ein Rechtssystem bilden.

Also gewährleistet $ {\begin{vmatrix}{\vec {a}}&{\vec {b}}&{\vec {c}}\end{vmatrix}}>0 $, dass die Vektoren $ {\vec {a}},{\vec {b}},{\vec {c}} $ eine rechtshändige Basis bilden.

Die Spaltenvektoren bilden eine Orthonormalbasis, wenn

$ M^{\top }M={\begin{pmatrix}1&0&0\\0&1&0\\0&0&1\end{pmatrix}} $

worin $ M^{\top } $ die transponierte Matrix ist. Bei der hier vorausgesetzten Rechtshändigkeit gilt dann zusätzlich $ |M|=+1 $.

Vektoralgebra

Basis und Duale Basis

Basisvektoren $ {\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3} $

Duale Basisvektoren $ {\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3} $

Beziehungen zwischen den Basisvektoren

$ {\vec {g}}_{i}\cdot {\vec {g}}^{j}=\delta _{i}^{j} $
$ {\vec {g}}^{1}={\frac {{\vec {g}}_{2}\times {\vec {g}}_{3}}{({\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3})}},\quad g^{2}={\frac {{\vec {g}}_{3}\times {\vec {g}}_{1}}{({\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3})}},\quad g^{3}={\frac {{\vec {g}}_{1}\times {\vec {g}}_{2}}{({\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3})}} $
$ {\vec {g}}_{1}={\frac {{\vec {g}}^{2}\times {\vec {g}}^{3}}{({\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3})}},\quad g_{2}={\frac {{\vec {g}}^{3}\times {\vec {g}}^{1}}{({\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3})}},\quad g_{3}={\frac {{\vec {g}}^{1}\times {\vec {g}}^{2}}{({\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3})}} $

mit dem Spatprodukt

$ ({\vec {a}},{\vec {b}},{\vec {c}}):={\vec {a}}\cdot ({\vec {b}}\times {\vec {c}})={\vec {c}}\cdot ({\vec {a}}\times {\vec {b}})={\vec {b}}\cdot ({\vec {c}}\times {\vec {a}})={\begin{vmatrix}{\vec {a}}&{\vec {b}}&{\vec {c}}\end{vmatrix}} $

Trägt man die Basisvektoren spaltenweise in eine Matrix ein, dann finden sich die dualen Basisvektoren in den Zeilen der Inversen oder den Spalten der #transponiert #Inversen $ ()^{\top -1} $:

$ {\begin{pmatrix}{\vec {g}}^{1}&{\vec {g}}^{2}&{\vec {g}}^{3}\end{pmatrix}}={\begin{pmatrix}{\vec {g}}_{1}&{\vec {g}}_{2}&{\vec {g}}_{3}\end{pmatrix}}^{\top -1} $

In der Standardbasis wie in jeder Orthonormalbasis sind die Basisvektoren $ {\hat {e}}_{1},{\hat {e}}_{2},{\hat {e}}_{3} $ zu sich selbst dual:

$ {\hat {e}}_{i}={\hat {e}}^{i} $

Berechnung von Vektorkomponenten

$ {\vec {v}}=v_{i}{\hat {e}}_{i}\quad \rightarrow \;v_{i}={\vec {v}}\cdot {\hat {e}}_{i} $
$ {\vec {v}}=v^{i}{\vec {g}}_{i}\quad \rightarrow \;v^{i}={\vec {v}}\cdot {\vec {g}}^{i} $
$ {\vec {v}}=v_{i}{\vec {g}}^{i}\quad \rightarrow \;v_{i}={\vec {v}}\cdot {\vec {g}}_{i} $

Beziehung zwischen den Skalarprodukten der Basisvektoren

$ ({\vec {g}}_{i}\cdot {\vec {g}}_{k})({\vec {g}}^{k}\cdot {\vec {g}}^{j})={\vec {g}}_{i}\cdot ({\vec {g}}^{j}\cdot {\vec {g}}^{k}){\vec {g}}_{k}={\vec {g}}_{i}\cdot {\vec {g}}^{j}=\delta _{i}^{j} $

Wechsel der Basis bei Vektoren

Wechsel von

Basis $ {\vec {g}}^{1},{\vec {g}}^{2},{\vec {g}}^{3} $ mit dualer Basis $ {\vec {g}}_{1},{\vec {g}}_{2},{\vec {g}}_{3} $

nach

Basis $ {\vec {h}}^{1},{\vec {h}}^{2},{\vec {h}}^{3} $ mit dualer Basis $ {\vec {h}}_{1},{\vec {h}}_{2},{\vec {h}}_{3} $:

$ {\vec {v}}=v_{i}\,{\vec {g}}^{i}=v_{i}^{\ast }\,{\vec {h}}^{i}\quad \rightarrow \;v_{i}^{\ast }=({\vec {h}}_{i}\cdot {\vec {g}}^{j})v_{j} $

Matrizengleichung:

$ {\begin{aligned}{\begin{pmatrix}v_{1}^{\ast }\\v_{2}^{\ast }\\v_{3}^{\ast }\end{pmatrix}}=&{\begin{pmatrix}{\vec {h}}_{1}\cdot {\vec {g}}^{1}&{\vec {h}}_{1}\cdot {\vec {g}}^{2}&{\vec {h}}_{1}\cdot {\vec {g}}^{3}\\{\vec {h}}_{2}\cdot {\vec {g}}^{1}&{\vec {h}}_{2}\cdot {\vec {g}}^{2}&{\vec {h}}_{2}\cdot {\vec {g}}^{3}\\{\vec {h}}_{3}\cdot {\vec {g}}^{1}&{\vec {h}}_{3}\cdot {\vec {g}}^{2}&{\vec {h}}_{3}\cdot {\vec {g}}^{3}\end{pmatrix}}{\begin{pmatrix}v_{1}\\v_{2}\\v_{3}\end{pmatrix}}\\=&{\begin{pmatrix}{\vec {h}}_{1}&{\vec {h}}_{2}&{\vec {h}}_{3}\end{pmatrix}}^{\top }{\begin{pmatrix}{\vec {g}}^{1}&{\vec {g}}^{2}&{\vec {g}}^{3}\end{pmatrix}}{\begin{pmatrix}v_{1}\\v_{2}\\v_{3}\end{pmatrix}}\end{aligned}} $

Dyadisches Produkt

Die grundlegenden Eigenschaften des dyadischen Produkts „⊗“ sind:

Abbildung $ \mathbb {V} \times \mathbb {V} \to {\mathcal {L}} $

$ {\vec {a}}\otimes {\vec {g}}=\mathbf {T} \in {\mathcal {L}} $

Multiplikation mit einem Skalar:

$ x({\vec {a}}\otimes {\vec {g}})=(x{\vec {a}})\otimes {\vec {g}}={\vec {a}}\otimes (x{\vec {g}})=x{\vec {a}}\otimes {\vec {g}} $

Distributivität:

$ (x+y){\vec {a}}\otimes {\vec {g}}=x{\vec {a}}\otimes {\vec {g}}+y{\vec {a}}\otimes {\vec {g}} $
$ ({\vec {a}}+{\vec {b}})\otimes {\vec {g}}={\vec {a}}\otimes {\vec {g}}+{\vec {b}}\otimes {\vec {g}} $
$ {\vec {a}}\otimes ({\vec {g}}+{\vec {h}})={\vec {a}}\otimes {\vec {g}}+{\vec {a}}\otimes {\vec {h}} $

Skalarprodukt:

$ ({\vec {a}}\otimes {\vec {g}}):({\vec {b}}\otimes {\vec {h}})=({\vec {a}}\cdot {\vec {b}})({\vec {g}}\cdot {\vec {h}}) $

Weitere Eigenschaften von Dyaden siehe #Dyade und den folgenden Abschnitt.

Tensoren als Elemente eines Vektorraumes

Durch die Eigenschaften des dyadischen Produktes wird $ {\mathcal {L}} $ zu einem euklidischen Vektorraum und entsprechend kann jeder Tensor komponentenweise bezüglich einer Basis von $ {\mathcal {L}} $ dargestellt werden:

$ \mathbf {A} \in {\mathcal {L}}\rightarrow \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}=A^{ij}{\vec {a}}_{i}\otimes {\vec {g}}_{j} $ mit Komponenten $ A_{ij},A^{ij}\in \mathbb {R} $.

Die Dyaden $ \{{\hat {e}}_{i}\otimes {\hat {e}}_{j}|i,j=1,2,3\} $ und $ \{{\vec {a}}_{i}\otimes {\vec {g}}_{j}|i,j=1,2,3\} $ bilden Basissysteme von $ {\mathcal {L}} $.

Operatoren

Transposition

Abbildung $ {\mathcal {L}}\to {\mathcal {L}} $

$ ({\vec {a}}\otimes {\vec {g}})^{\top }:={\vec {g}}\otimes {\vec {a}} $
$ (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})^{\top }=A_{ij}({\hat {e}}_{j}\otimes {\hat {e}}_{i})=A_{ji}({\hat {e}}_{i}\otimes {\hat {e}}_{j}) $
$ (A^{ij}{\vec {a}}_{i}\otimes {\vec {g}}_{j})^{\top }=A^{ij}({\vec {g}}_{j}\otimes {\vec {a}}_{i})=A^{ji}({\vec {g}}_{i}\otimes {\vec {a}}_{j}) $
$ \left(\mathbf {A} ^{\top }\right)^{\top }=\mathbf {A} $
$ (\mathbf {A+B} )^{\top }=\mathbf {A} ^{\top }+\mathbf {B} ^{\top } $
$ (\mathbf {A\cdot B} )^{\top }=\mathbf {B} ^{\top }\cdot \mathbf {A} ^{\top } $

Vektortransformation

Abbildung $ {\mathcal {L}}\times \mathbb {V} \to \mathbb {V} $ oder $ \mathbb {V} \times {\mathcal {L}}\to \mathbb {V} $

Dyaden:

$ ({\vec {a}}\otimes {\vec {g}})\cdot {\vec {h}}:=({\vec {g}}\cdot {\vec {h}}){\vec {a}} $
$ {\vec {b}}\cdot ({\vec {a}}\otimes {\vec {g}}):=({\vec {a}}\cdot {\vec {b}}){\vec {g}} $
$ ({\vec {a}}\otimes {\vec {g}})\cdot {\vec {h}}={\vec {h}}\cdot ({\vec {a}}\otimes {\vec {g}})^{\top } $
$ {\vec {b}}\cdot ({\vec {a}}\otimes {\vec {g}})=({\vec {a}}\otimes {\vec {g}})^{\top }\cdot {\vec {b}} $

Allgemeine Tensoren:

$ A_{ij}({\hat {e}}_{i}\otimes {\hat {e}}_{j})\cdot {\vec {v}}=A_{ij}({\vec {v}}\cdot {\hat {e}}_{j}){\hat {e}}_{i} $
$ A^{ij}({\vec {a}}_{i}\otimes {\vec {g}}_{j})\cdot {\vec {v}}=A^{ij}({\vec {v}}\cdot {\vec {g}}_{j}){\vec {a}}_{i} $
$ {\vec {v}}\cdot A_{ij}({\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{ij}({\vec {v}}\cdot {\hat {e}}_{i}){\hat {e}}_{j} $
$ {\vec {v}}\cdot A^{ij}({\vec {a}}_{i}\otimes {\vec {g}}_{j})=A^{ij}({\vec {v}}\cdot {\hat {a}}_{i}){\vec {g}}_{j} $

Symbolisch:

$ \mathbf {A} \cdot {\vec {v}}={\vec {v}}\cdot \mathbf {A} ^{\top } $
$ {\vec {v}}\cdot \mathbf {A} =\mathbf {A} ^{\top }\cdot {\vec {v}} $

Tensorprodukt

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to {\mathcal {L}} $

$ ({\vec {a}}\otimes {\vec {g}})\cdot ({\vec {h}}\otimes {\vec {u}}):=({\vec {g}}\cdot {\vec {h}}){\vec {a}}\otimes {\vec {u}} $
$ ({\vec {a}}\otimes {\vec {g}})\cdot \mathbf {A} ={\vec {a}}\otimes ({\vec {g}}\cdot \mathbf {A} )={\vec {a}}\otimes {\vec {g}}\cdot \mathbf {A} ={\vec {a}}\otimes (\mathbf {A} ^{\top }\cdot {\vec {g}}) $
$ \mathbf {A} \cdot ({\vec {a}}\otimes {\vec {g}})=(\mathbf {A} \cdot {\vec {a}})\otimes {\vec {g}}=\mathbf {A} \cdot {\vec {a}}\otimes {\vec {g}} $
$ (A_{ik}{\hat {e}}_{i}\otimes {\hat {e}}_{k})\cdot (B_{lj}{\hat {e}}_{l}\otimes {\hat {e}}_{j})=A_{ik}B_{kj}{\hat {e}}_{i}\otimes {\hat {e}}_{j} $
$ \left(A^{ij}{\vec {a}}_{i}\otimes {\vec {g}}_{j}\right)\cdot \left(B^{kl}{\vec {h}}_{k}\otimes {\vec {u}}_{l}\right)=A^{ij}({\vec {g}}_{j}\cdot {\vec {h}}_{k})B^{kl}{\vec {a}}_{i}\otimes {\vec {u}}_{l} $

Skalarprodukt von Tensoren

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to \mathbb {R} $

Definition über die #Spur:

$ ({\vec {a}}\otimes {\vec {g}}):({\vec {b}}\otimes {\vec {h}}):=\mathrm {Sp} (({\vec {a}}\otimes {\vec {g}})^{\top }\cdot ({\vec {b}}\otimes {\vec {h}}))=({\vec {a}}\cdot {\vec {b}})({\vec {g}}\cdot {\vec {h}}) $
$ \mathbf {A} :\mathbf {B} :=\mathrm {Sp} (\mathbf {A} ^{\top }\cdot \mathbf {B} ) $

Eigenschaften:

$ \mathbf {A} :\mathbf {B} =\mathbf {B} :\mathbf {A} =\mathbf {A} ^{\top }:\mathbf {B} ^{\top }=\mathbf {B} ^{\top }:\mathbf {A} ^{\top } $
$ \mathbf {A} ^{\top }:\mathbf {B} =\mathbf {A} :\mathbf {B} ^{\top } $
$ \mathbf {A} :(\mathbf {B\cdot C} )=(\mathbf {B} ^{\top }\cdot \mathbf {A} ):\mathbf {C} =(\mathbf {A\cdot C} ^{\top }):\mathbf {B} $
$ (\mathbf {A\cdot B} ):\mathbf {C} =\mathbf {B} :(\mathbf {A} ^{\top }\cdot \mathbf {C} )=\mathbf {A} :(\mathbf {C\cdot B} ^{\top }) $
$ ({\vec {u}}\otimes {\vec {v}}):\mathbf {A} ={\vec {u}}\cdot \mathbf {A} \cdot {\vec {v}} $

Kreuzprodukt eines Vektors mit einem Tensor

Abbildung $ \mathbb {V} \times {\mathcal {L}}\to {\mathcal {L}} $ oder $ {\mathcal {L}}\times \mathbb {V} \to {\mathcal {L}} $

Dyaden:

$ {\vec {a}}\times ({\vec {b}}\otimes {\vec {g}})=({\vec {a}}\times {\vec {b}})\otimes {\vec {g}}={\vec {a}}\times {\vec {b}}\otimes {\vec {g}} $
$ ({\vec {a}}\otimes {\vec {g}})\times {\vec {h}}={\vec {a}}\otimes ({\vec {g}}\times {\vec {h}})={\vec {a}}\otimes {\vec {g}}\times {\vec {h}} $
$ {\vec {a}}\times {\vec {b}}\otimes {\vec {g}}=-[({\vec {b}}\otimes {\vec {g}})^{\top }\times {\vec {a}}]^{\top } $
$ {\vec {a}}\otimes {\vec {g}}\times {\vec {h}}=-[{\vec {h}}\times ({\vec {a}}\otimes {\vec {g}})^{\top }]^{\top } $
$ a_{j}{\hat {e}}_{j}\times (A_{kl}{\hat {e}}_{k}\otimes {\hat {e}}_{l})=a_{j}A_{kl}({\hat {e}}_{j}\times {\hat {e}}_{k})\otimes {\hat {e}}_{l}=\epsilon _{ijk}a_{j}A_{kl}{\hat {e}}_{i}\otimes {\hat {e}}_{l} $
$ (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})\times a_{k}{\hat {e}}_{k}=A_{ij}a_{k}{\hat {e}}_{i}\otimes ({\hat {e}}_{j}\times {\hat {e}}_{k})=\epsilon _{jkl}A_{ij}a_{k}{\hat {e}}_{i}\otimes {\hat {e}}_{l} $

Allgemeine Tensoren:

$ ({\vec {a}}\times \mathbf {A} )\cdot {\vec {g}}:={\vec {a}}\times (\mathbf {A} \cdot {\vec {g}})={\vec {a}}\times ({\vec {g}}\cdot \mathbf {A} ^{\top }) $
$ {\vec {b}}\cdot ({\vec {a}}\times \mathbf {A} ):=({\vec {b}}\times {\vec {a}})\cdot \mathbf {A} $
$ {\vec {g}}\cdot (\mathbf {A} \times {\vec {a}}):=({\vec {g}}\cdot \mathbf {A} )\times {\vec {a}}=(\mathbf {A} ^{\top }\cdot {\vec {g}})\times {\vec {a}} $
$ (\mathbf {A} \times {\vec {a}})\cdot {\vec {b}}=\mathbf {A} \cdot ({\vec {a}}\times {\vec {b}}) $
$ {\vec {a}}\times \mathbf {A} =-\left(\mathbf {A} ^{\top }\times {\vec {a}}\right)^{\top } $
$ \mathbf {A} \times {\vec {a}}=-\left({\vec {a}}\times \mathbf {A} ^{\top }\right)^{\top } $

Symmetrische Tensoren: $ {\vec {a}}\times \mathbf {A} ^{\mathrm {S} }=-\left(\mathbf {A} ^{\mathrm {S} }\times {\vec {a}}\right)^{\top } $

Insbesondere Kugeltensoren: $ {\vec {a}}\times \mathbf {A} ^{\mathrm {K} }=\mathbf {A} ^{\mathrm {K} }\times {\vec {a}}=-({\vec {a}}\times \mathbf {A} ^{\mathrm {K} })^{\top } $

Schiefsymmetrische Tensoren: $ {\vec {a}}\times \mathbf {A} ^{\mathrm {A} }=\left(\mathbf {A} ^{\mathrm {A} }\times {\vec {a}}\right)^{\top } $

#Axialer Tensor oder Kreuzproduktmatrix mit dem #Einheitstensor:

$ ({\vec {a}}\times \mathbf {1} )\cdot {\vec {g}}={\vec {a}}\cdot ({\vec {g}}\times \mathbf {1} )={\vec {a}}\cdot (\mathbf {1} \times {\vec {g}})={\vec {a}}\times {\vec {g}} $

Mehrfach:

$ ({\vec {a}}\times ({\vec {b}}\times \mathbf {A} ))\cdot {\vec {g}}={\vec {a}}\times ({\vec {b}}\times (\mathbf {A} \cdot {\vec {g}}))=({\vec {a}}\cdot \mathbf {A} \cdot {\vec {g}}){\vec {b}}-({\vec {a}}\cdot {\vec {b}})\mathbf {A} \cdot {\vec {g}} $
$ {\vec {a}}\times ({\vec {b}}\times \mathbf {A} )={\vec {b}}\otimes {\vec {a}}\cdot \mathbf {A} -({\vec {a}}\cdot {\vec {b}})\mathbf {A} $

Meistens ist aber:

$ (\mathbf {A} \cdot {\vec {a}})\times {\vec {g}}\neq \mathbf {A} \cdot ({\vec {a}}\times {\vec {g}})=(\mathbf {A} \times {\vec {a}})\cdot {\vec {g}} $
$ {\vec {a}}\times ({\vec {g}}\cdot \mathbf {A} )\neq ({\vec {a}}\times {\vec {g}})\cdot \mathbf {A} ={\vec {a}}\cdot ({\vec {g}}\times \mathbf {A} ) $

Kreuzprodukt von Tensoren

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to \mathbb {V} $

$ \mathbf {A\times B} ={\stackrel {3}{\mathbf {E} }}:(\mathbf {A\cdot B} ^{\top })=-{\stackrel {3}{\mathbf {E} }}:(\mathbf {B\cdot A} ^{\top })=-\mathbf {B\times A} \in \mathbb {V} $

mit #Fundamentaltensor 3. Stufe $ {\stackrel {3}{\mathbf {E} }} $.

$ ({\vec {a}}\otimes {\vec {g}})\times ({\vec {b}}\otimes {\vec {h}})=({\vec {g}}\cdot {\vec {h}}){\vec {a}}\times {\vec {b}} $
$ {\begin{aligned}&A_{ik}({\hat {e}}_{i}\otimes {\hat {e}}_{k})\times [B_{jl}({\hat {e}}_{j}\otimes {\hat {e}}_{l})]=A_{ik}B_{jk}({\hat {e}}_{i}\times {\hat {e}}_{j})=\ldots \\&\ldots ={\begin{pmatrix}A_{21}B_{31}-A_{31}B_{21}+A_{22}B_{32}-A_{32}B_{22}+A_{23}B_{33}-A_{33}B_{23}\\A_{31}B_{11}-A_{11}B_{31}+A_{32}B_{12}-A_{12}B_{32}+A_{33}B_{13}-A_{13}B_{33}\\A_{11}B_{21}-A_{21}B_{11}+A_{12}B_{22}-A_{22}B_{12}+A_{13}B_{23}-A_{23}B_{13}\end{pmatrix}}\end{aligned}} $

Zusammenhang mit #Dualer axialer Vektor und #Vektorinvariante:

$ \mathbf {A\times B} =-2{\stackrel {A}{\overrightarrow {\mathbf {A\cdot B} ^{\top }}}}={\vec {\mathrm {i} }}(\mathbf {A\cdot B} ^{\top }) $

Mit #Einheitstensor:

$ \mathbf {1\times A} =2{\stackrel {A}{\overrightarrow {\mathbf {A} }}}=-{\vec {\mathrm {i} }}(\mathbf {A} ) $

Mehrfachprodukte:

$ (\mathbf {A\cdot B} )\times \mathbf {C} =\mathbf {A} \times (\mathbf {C\cdot B} ^{\top }) $
$ \mathbf {A} \times (\mathbf {B\cdot C} )=(\mathbf {A\cdot C} ^{\top })\times \mathbf {B} $

Zusammenhang mit dem #Skalarkreuzprodukt von Tensoren:

$ \mathbf {A\times B} =\mathbf {A} \cdot \!\!\times (\mathbf {B} ^{\top }) $

Skalarkreuzprodukt von Tensoren

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to \mathbb {V} $

$ ({\vec {a}}\otimes {\vec {g}})\cdot \!\!\times ({\vec {h}}\otimes {\vec {u}})=-({\vec {u}}\otimes {\vec {h}})\cdot \!\!\times ({\vec {g}}\otimes {\vec {a}}):=({\vec {g}}\cdot {\vec {h}}){\vec {a}}\times {\vec {u}} $
$ {\begin{aligned}&A_{ik}({\hat {e}}_{i}\otimes {\hat {e}}_{k})\cdot \!\!\times [B_{lj}({\hat {e}}_{l}\otimes {\hat {e}}_{j})]=A_{ik}B_{kj}({\hat {e}}_{i}\times {\hat {e}}_{j})=\ldots \\&\ldots ={\begin{pmatrix}A_{21}B_{13}-A_{31}B_{12}+A_{22}B_{23}-A_{32}B_{22}+A_{23}B_{33}-A_{33}B_{32}\\A_{31}B_{11}-A_{11}B_{13}+A_{32}B_{21}-A_{12}B_{23}+A_{33}B_{31}-A_{13}B_{33}\\A_{11}B_{12}-A_{21}B_{11}+A_{12}B_{22}-A_{22}B_{21}+A_{13}B_{32}-A_{23}B_{31}\end{pmatrix}}\end{aligned}} $

Das Skalarkreuzprodukt mit dem #Einheitstensor vertauscht das dyadische Produkt durch das Kreuzprodukt:

$ \mathbf {1} \cdot \!\!\times ({\vec {a}}\otimes {\vec {b}})={\vec {a}}\times {\vec {b}} $

Allgemein:

$ \mathbf {A} \cdot \!\!\times \mathbf {B} =-(\mathbf {B} ^{\top })\cdot \!\!\times (\mathbf {A} ^{\top }) $
$ \mathbf {A} \cdot \!\!\times (\mathbf {B\cdot C} )=(\mathbf {A\cdot B} )\cdot \!\!\times \mathbf {C} $
$ (\mathbf {A\cdot B} )\cdot \!\!\times \mathbf {C} =\mathbf {A} \cdot \!\!\times (\mathbf {B\cdot C} ) $

Zusammenhang mit dem #Kreuzprodukt von Tensoren:

$ \mathbf {S} \cdot \!\!\times \mathbf {T} =\mathbf {S\times (T^{\top })} $

Zusammenhang mit #Vektorinvariante und #Dualer axialer Vektor:

$ \mathbf {A} \cdot \!\!\times \mathbf {B} ={\vec {\mathrm {i} }}(\mathbf {A} \cdot \mathbf {B} )=-2{\stackrel {A}{\overrightarrow {\mathbf {A} \cdot \mathbf {B} }}} $

Doppeltes Kreuzprodukt von Tensoren

Siehe auch #Äußeres Tensorprodukt #

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to {\mathcal {L}} $

$ ({\vec {a}}\otimes {\vec {g}})\times \!\!\times ({\vec {h}}\otimes {\vec {b}}):=({\vec {g}}\times {\vec {h}})\otimes ({\vec {a}}\times {\vec {b}})=({\vec {g}}\otimes {\vec {a}})\#({\vec {h}}\otimes {\vec {b}}) $
$ A_{ij}({\hat {e}}_{i}\otimes {\hat {e}}_{j})\times \!\!\times [B_{kl}({\hat {e}}_{k}\otimes {\hat {e}}_{l})]:=A_{ij}B_{kl}({\hat {e}}_{j}\times {\hat {e}}_{k})\otimes ({\hat {e}}_{i}\times {\hat {e}}_{l}) $
$ \mathbf {A} \times \!\!\times \mathbf {B} =\mathbf {A} ^{\top }\#\mathbf {B} $

Äußeres Tensorprodukt

Abbildung $ {\mathcal {L}}\times {\mathcal {L}}\to {\mathcal {L}} $

$ ({\vec {a}}\otimes {\vec {g}})\#({\vec {b}}\otimes {\vec {h}}):=({\vec {a}}\times {\vec {b}})\otimes ({\vec {g}}\times {\vec {h}})=({\vec {g}}\otimes {\vec {a}})\times \!\!\times ({\vec {b}}\otimes {\vec {h}}) $
$ {\begin{aligned}&(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})\#(B_{kl}{\hat {e}}_{k}\otimes {\hat {e}}_{l})=A_{ij}B_{kl}({\hat {e}}_{i}\times {\hat {e}}_{k})\otimes ({\hat {e}}_{j}\times {\hat {e}}_{l})\\&\qquad \qquad \qquad \qquad \qquad \quad \;\;\;=\epsilon _{ikm}\epsilon _{jln}A_{ij}B_{kl}{\hat {e}}_{m}\otimes {\hat {e}}_{n}\end{aligned}} $

Mit der Formel für das Produkt zweier #Permutationssymbole:

$ {\begin{aligned}\mathbf {A} \#\mathbf {B} =&[\mathrm {Sp} (\mathbf {A} )\mathrm {Sp} (\mathbf {B} )-\mathrm {Sp} (\mathbf {A\cdot B} )]\mathbf {1} \\&+[\mathbf {A\cdot B} +\mathbf {B\cdot A} -\mathrm {Sp} (\mathbf {A} )\mathbf {B} -\mathrm {Sp} (\mathbf {B} )\mathbf {A} ]^{\top }\end{aligned}} $

Grundlegende Eigenschaften:

$ \mathbf {A} \#\mathbf {B} =\mathbf {B} \#\mathbf {A} =(\mathbf {A} ^{\top }\#\mathbf {B} ^{\top })^{\top } $
$ (\mathbf {A+B} )\#\mathbf {C} =\mathbf {A} \#\mathbf {C} +\mathbf {B} \#\mathbf {C} $
$ \mathbf {A} \#(\mathbf {B+C} )=\mathbf {A} \#\mathbf {B} +\mathbf {A} \#\mathbf {C} $

Kreuzprodukt und #Kofaktor:

$ (\mathbf {A} \#\mathbf {B} )\cdot ({\vec {u}}\times {\vec {v}})=(\mathbf {A} \cdot {\vec {u}})\times (\mathbf {B} \cdot {\vec {v}})-(\mathbf {A} \cdot {\vec {v}})\times (\mathbf {B} \cdot {\vec {u}}) $
$ {\frac {1}{2}}(\mathbf {A} \#\mathbf {A} )\cdot ({\vec {u}}\times {\vec {v}})=\mathrm {cof} (\mathbf {A} )\cdot ({\vec {u}}\times {\vec {v}})=(\mathbf {A} \cdot {\vec {u}})\times (\mathbf {A} \cdot {\vec {v}}) $

#Hauptinvarianten:

$ {\frac {1}{2}}(\mathbf {A\#1} ):\mathbf {1} =\mathrm {Sp} (\mathbf {A} ) $
$ {\frac {1}{2}}(\mathbf {A\#A} ):\mathbf {1} =\mathrm {I} _{2}(\mathbf {A} ) $
$ {\frac {1}{6}}(\mathbf {A\#A} ):\mathbf {A} =\det(\mathbf {A} ) $

Weitere Eigenschaften:

$ \mathbf {1} \#\mathbf {1} =2\,\mathbf {1} $
$ \mathbf {A} \#\mathbf {1} =\mathrm {Sp} (\mathbf {A} )\mathbf {1} -\mathbf {A} ^{\top } $
$ (\mathbf {A} \#\mathbf {B} ):\mathbf {C} =(\mathbf {B} \#\mathbf {C} ):\mathbf {A} =(\mathbf {C} \#\mathbf {A} ):\mathbf {B} $
$ \mathrm {Sp} (\mathbf {A} \#\mathbf {B} )=\mathrm {Sp} (\mathbf {A} )\mathrm {Sp} (\mathbf {B} )-\mathrm {Sp} (\mathbf {A\cdot B} ) $
$ (\mathbf {A} \#\mathbf {B} )\cdot (\mathbf {C} \#\mathbf {D} )=(\mathbf {A\cdot C} )\#(\mathbf {B\cdot D} )+(\mathbf {A\cdot D} )\#(\mathbf {B\cdot C} ) $

Aber meistens:

$ (\mathbf {A} \#\mathbf {B} )\#\mathbf {C} \neq \mathbf {A} \#(\mathbf {B} \#\mathbf {C} ) $

.

Produkte von Tensoren, Dyaden und Vektoren

$ \mathbf {A} \cdot ({\vec {a}}\otimes {\vec {g}})=(\mathbf {A} \cdot {\vec {a}})\otimes {\vec {g}} $
$ {\vec {a}}\otimes (\mathbf {A} \cdot {\vec {g}})=({\vec {a}}\otimes {\vec {g}})\cdot \mathbf {A} ^{\top } $
$ {\vec {a}}\cdot \mathbf {A} \cdot {\vec {g}}=\mathbf {A} :({\vec {a}}\otimes {\vec {g}}) $

Spatprodukt und #Determinante eines Tensors:

$ (\mathbf {A} \cdot {\vec {a}})\cdot [(\mathbf {A} \cdot {\vec {b}})\times (\mathbf {A} \cdot {\vec {c}})]=\mathrm {det} (\mathbf {A} )\;{\vec {a}}\cdot ({\vec {b}}\times {\vec {c}}) $

Kreuzprodukt und #Kofaktor:

$ (\mathbf {A} \cdot {\vec {a}})\times (\mathbf {A} \cdot {\vec {b}})=\mathrm {cof} (\mathbf {A} )\cdot ({\vec {a}}\times {\vec {b}}) $
$ \mathbf {A} ^{\top }\cdot [(\mathbf {A} \cdot {\vec {a}})\times (\mathbf {A} \cdot {\vec {b}})]=\mathrm {det} (\mathbf {A} )\;{\vec {a}}\times {\vec {b}} $

#Axialer Tensor oder Kreuzproduktmatrix, #Kreuzprodukt von Tensoren, #Skalarkreuzprodukt von Tensoren, #Dualer axialer Vektor und #Vektorinvariante:

$ ({\vec {u}}\times \mathbf {1} )\cdot {\vec {v}}=({\vec {u}}\otimes {\vec {v}})\times \mathbf {1} =({\vec {u}}\otimes {\vec {v}})\cdot \!\!\times \mathbf {1} ={\stackrel {A}{\overrightarrow {({\vec {u}}\times {\vec {v}})\times \mathbf {1} }}}={\vec {\mathrm {i} }}({\vec {u}}\otimes {\vec {v}})={\vec {u}}\times {\vec {v}} $

Tensorkomponenten

$ \mathbf {A} =A_{ij}\,{\hat {e}}_{i}\otimes {\hat {e}}_{j}={\begin{pmatrix}A_{11}&A_{12}&A_{13}\\A_{21}&A_{22}&A_{23}\\A_{31}&A_{32}&A_{33}\end{pmatrix}}\quad \rightarrow \;A_{ij}={\hat {e}}_{i}\cdot \mathbf {A} \cdot {\hat {e}}_{j} $
$ \mathbf {A} =A^{ij}\,{\vec {a}}_{i}\otimes {\vec {g}}_{j}\quad \rightarrow \;A^{ij}={\vec {a}}^{i}\cdot \mathbf {A} \cdot {\vec {g}}^{j}=({\vec {a}}^{i}\otimes {\vec {g}}^{j}):\mathbf {A} $
$ \mathbf {A} =A_{ij}\,{\vec {a}}^{i}\otimes {\vec {g}}^{j}\quad \rightarrow \;A_{ij}={\vec {a}}_{i}\cdot \mathbf {A} \cdot {\vec {g}}_{j} $
$ \mathbf {A} =A_{j}^{i}\,{\vec {a}}_{i}\otimes {\vec {g}}^{j}\quad \rightarrow \;A_{j}^{i}={\vec {a}}^{i}\cdot \mathbf {A} \cdot {\vec {g}}_{j} $
$ \mathbf {A} =A_{i}^{j}\,{\vec {a}}^{i}\otimes {\vec {g}}_{j}\quad \rightarrow \;A_{i}^{j}={\vec {a}}_{i}\cdot \mathbf {A} \cdot {\vec {g}}^{j} $

Wechsel der Basis

$ \mathbf {A} =A_{ij}{\vec {a}}^{i}\otimes {\vec {a}}^{j}=A_{ij}^{\ast }{\vec {b}}^{i}\otimes {\vec {b}}^{j} $

Die Komponenten $ A_{ij}^{\ast } $ ergeben sich durch Vor- und Nachmultiplikation mit dem #Einheitstensor $ \mathbf {1} ={\vec {b}}^{i}\otimes {\vec {b}}_{i} $:

$ {\begin{aligned}\mathbf {A} =\mathbf {1\cdot A\cdot 1} ^{\top }=&({\vec {b}}^{i}\otimes {\vec {b}}_{i})\cdot (A_{kl}{\vec {a}}^{k}\otimes {\vec {a}}^{l})\cdot ({\vec {b}}_{j}\otimes {\vec {b}}^{j})\\=&({\vec {b}}_{i}\cdot {\vec {a}}^{k})A_{kl}({\vec {a}}^{l}\cdot {\vec {b}}_{j}){\vec {b}}^{i}\otimes {\vec {b}}^{j}=:A_{ij}^{\ast }{\vec {b}}^{i}\otimes {\vec {b}}^{j}\\\rightarrow A_{ij}^{\ast }=&({\vec {b}}_{i}\cdot {\vec {a}}^{k})A_{kl}({\vec {a}}^{l}\cdot {\vec {b}}_{j})\end{aligned}} $

Allgemein:

$ \mathbf {A} =A_{ij}{\vec {a}}^{i}\otimes {\vec {g}}^{j}=A_{ij}^{\ast }{\vec {b}}^{i}\otimes {\vec {h}}^{j} $

Basiswechsel mit $ \mathbf {1} =({\vec {b}}_{i}\cdot {\vec {a}}^{k}){\vec {b}}^{i}\otimes {\vec {a}}_{k}=({\vec {h}}_{j}\cdot {\vec {g}}^{l}){\vec {h}}^{j}\otimes {\vec {g}}_{l} $:

$ {\begin{aligned}\mathbf {A} =\mathbf {1\cdot A\cdot 1} ^{\top }=&({\vec {b}}_{i}\cdot {\vec {a}}^{k})({\vec {b}}^{i}\otimes {\vec {a}}_{k})\cdot A_{mn}({\vec {a}}^{m}\otimes {\vec {g}}^{n})\cdot ({\vec {h}}_{j}\cdot {\vec {g}}^{l})({\vec {g}}_{l}\otimes {\vec {h}}^{j})\\=&({\vec {b}}_{i}\cdot {\vec {a}}^{k})A_{kl}({\vec {h}}_{j}\cdot {\vec {g}}^{l})({\vec {b}}^{i}\otimes {\vec {h}}^{j})=A_{ij}^{\ast }{\vec {b}}^{i}\otimes {\vec {h}}^{j}\\\rightarrow A_{ij}^{\ast }=&({\vec {b}}_{i}\cdot {\vec {a}}^{k})A_{kl}({\vec {g}}^{l}\cdot {\vec {h}}_{j})\end{aligned}} $

Bilinearform und Identität von Tensoren

Definition für einen Tensor A:

$ \langle {\vec {u}},{\vec {v}}\rangle :={\vec {u}}\cdot \mathbf {A} \cdot {\vec {v}}=\mathbf {A} :({\vec {u}}\otimes {\vec {v}}) $

Zwei Tensoren A und B sind identisch, wenn

$ {\vec {u}}\cdot \mathbf {A} \cdot {\vec {v}}={\vec {u}}\cdot \mathbf {B} \cdot {\vec {v}}\quad \forall \;{\vec {u}},{\vec {v}}\in \mathbb {V} $

Kofaktor

Definition

$ \mathrm {cof} (\mathbf {A} ):=\mathbf {A^{\top }\cdot A^{\top }} -\mathrm {I} _{1}(\mathbf {A} )\mathbf {A} ^{\top }+\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} $

#Invarianten:

Wenn λ1,2,3 die #Eigenwerte des Tensors A sind, dann hat cof(A) die Eigenwerte λ1λ2, λ2λ3, λ3λ1.

#Hauptinvarianten:

$ \mathrm {I} _{1}(\mathrm {cof} (\mathbf {A} ))=\mathrm {I} _{2}(\mathbf {A} ) $
$ \mathrm {I} _{2}(\mathrm {cof} (\mathbf {A} ))=\mathrm {I} _{1}(\mathbf {A} )\mathrm {det} (\mathbf {A} ) $
$ \mathrm {det} (\mathrm {cof} (\mathbf {A} ))=\mathrm {det} ^{2}(\mathbf {A} ) $

#Betrag:

$ \|\mathrm {cof} (\mathbf {A} )\|={\sqrt {\mathrm {I} _{2}(\mathbf {A^{\top }\cdot A} )}}={\frac {\sqrt {2}}{2}}{\sqrt {\|\mathbf {A} \|^{4}-\|\mathbf {A^{\top }\cdot A} \|^{2}}} $

Weitere Eigenschaften:

$ \mathrm {cof} (x\mathbf {A} )=x^{2}\mathrm {cof} (\mathbf {A} ) $
$ \mathrm {det} (\mathbf {A} )\neq 0\quad \rightarrow \quad \mathrm {cof} (\mathbf {A} )=\det(\mathbf {A} )\mathbf {A} ^{\top -1} $
$ \mathbf {A} ^{\top }\cdot \mathrm {cof} (\mathbf {A} )=\mathrm {cof} (\mathbf {A} )\cdot \mathbf {A} ^{\top }=\mathrm {det} (\mathbf {A} )\mathbf {1} $
$ \mathrm {cof} (\mathbf {A\cdot B} )=\mathrm {cof} (\mathbf {A} )\cdot \mathrm {cof} (\mathbf {B} ) $
$ \mathrm {cof} (\mathbf {A} ^{\top })=\mathrm {cof} (\mathbf {A} )^{\top } $
$ \mathrm {cof} \left(\mathrm {cof} (\mathbf {A} )\right)=\mathrm {det} (\mathbf {A} )\mathbf {A} $
$ {\begin{aligned}&\mathrm {cof} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})={\frac {1}{2}}(A_{kl}A_{mn}\epsilon _{kmi}\epsilon _{lnj})({\hat {e}}_{i}\otimes {\hat {e}}_{j})=\ldots \\&\ldots ={\begin{pmatrix}A_{22}A_{33}-A_{23}A_{32}&A_{23}A_{31}-A_{21}A_{33}&A_{21}A_{32}-A_{22}A_{31}\\A_{32}A_{13}-A_{33}A_{12}&A_{33}A_{11}-A_{31}A_{13}&A_{31}A_{12}-A_{32}A_{11}\\A_{12}A_{23}-A_{13}A_{22}&A_{13}A_{21}-A_{11}A_{23}&A_{11}A_{22}-A_{12}A_{21}\end{pmatrix}}\end{aligned}} $

Kofaktor und #Äußeres Tensorprodukt:

$ \mathrm {cof} (\mathbf {A} )={\frac {1}{2}}\mathbf {A} \#\mathbf {A} $
$ {\begin{aligned}\mathrm {cof} (\mathbf {A+B} )=&{\frac {1}{2}}(\mathbf {A} \#\mathbf {A} +2\mathbf {A} \#\mathbf {B} +\mathbf {B} \#\mathbf {B} )\\=&\mathrm {cof} (\mathbf {A} )+\mathrm {cof} (\mathbf {B} )+\mathbf {A} \#\mathbf {B} \end{aligned}} $

Kreuzprodukt und Kofaktor:

$ (\mathbf {A} \cdot {\vec {a}})\times (\mathbf {A} \cdot {\vec {b}})=\mathrm {cof} (\mathbf {A} )\cdot ({\vec {a}}\times {\vec {b}}) $

Adjunkte

Definition:

$ \mathrm {adj} (\mathbf {A} ):=\mathbf {A} ^{2}-\mathrm {I} _{1}(\mathbf {A} )\mathbf {A} +\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} =\mathrm {cof} (\mathbf {A} )^{\top } $

#Hauptinvarianten:

$ \mathrm {I} _{1}(\mathrm {adj} (\mathbf {A} ))=\mathrm {I} _{2}(\mathbf {A} ) $
$ \mathrm {I} _{2}(\mathrm {adj} (\mathbf {A} ))=\mathrm {I} _{1}(\mathbf {A} )\mathrm {det} (\mathbf {A} ) $
$ \mathrm {det} (\mathrm {adj} (\mathbf {A} ))=\mathrm {det} ^{2}(\mathbf {A} ) $

#Betrag:

$ \|\mathrm {adj} (\mathbf {A} )\|={\sqrt {\mathrm {I} _{2}(\mathbf {A^{\top }\cdot A} )}}={\frac {\sqrt {2}}{2}}{\sqrt {\|\mathbf {A} \|^{4}-\|\mathbf {A^{\top }\cdot A} \|^{2}}} $

Weitere Eigenschaften:

$ \mathrm {adj} (x\mathbf {A} )=x^{2}\mathrm {adj} (\mathbf {A} ) $
$ \mathrm {det} (\mathbf {A} )\neq 0\quad \rightarrow \quad \mathrm {adj} (\mathbf {A} )=\det(\mathbf {A} )\mathbf {A} ^{-1} $
$ \mathbf {A} \cdot \mathrm {adj} (\mathbf {A} )=\mathrm {adj} (\mathbf {A} )\cdot \mathbf {A} =\mathrm {det} (\mathbf {A} )\mathbf {1} $
$ \mathrm {adj} (\mathbf {A\cdot B} )=\mathrm {adj} (\mathbf {B} )\cdot \mathrm {adj} (\mathbf {A} ) $
$ \mathrm {adj} (\mathbf {A} ^{\top })=\mathrm {adj} (\mathbf {A} )^{\top } $
$ {\begin{aligned}\mathrm {adj} (\mathbf {A+B} )=&{\frac {1}{2}}(\mathbf {A} \#\mathbf {A} +2\mathbf {A} \#\mathbf {B} +\mathbf {B} \#\mathbf {B} )^{\top }\\=&\mathrm {adj} (\mathbf {A} )+\mathrm {adj} (\mathbf {B} )+\mathbf {A} ^{\top }\#\mathbf {B} ^{\top }\end{aligned}} $
$ \mathrm {adj} \left(\mathrm {adj} (\mathbf {A} )\right)=\mathrm {det} (\mathbf {A} )\mathbf {A} $
$ {\begin{aligned}&\mathrm {adj} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})={\frac {1}{2}}(A_{kl}A_{mn}\epsilon _{kmj}\epsilon _{lni})({\hat {e}}_{i}\otimes {\hat {e}}_{j})=\ldots \\&\ldots ={\begin{pmatrix}A_{22}A_{33}-A_{23}A_{32}&A_{32}A_{13}-A_{33}A_{12}&A_{12}A_{23}-A_{13}A_{22}\\A_{23}A_{31}-A_{21}A_{33}&A_{33}A_{11}-A_{31}A_{13}&A_{13}A_{21}-A_{11}A_{23}\\A_{21}A_{32}-A_{22}A_{31}&A_{31}A_{12}-A_{32}A_{11}&A_{11}A_{22}-A_{12}A_{21}\end{pmatrix}}\end{aligned}} $

Inverse

Definition

$ \mathbf {A} ^{-1}:\quad \mathbf {A} ^{-1}\cdot \mathbf {A} =\mathbf {A\cdot A} ^{-1}=\mathbf {1} $

Die Inverse ist nur definiert, wenn $ |\mathbf {A} |=\mathrm {det} (\mathbf {A} )=\mathrm {I} _{3}(\mathbf {A} )\neq 0 $

Zusammenhang mit dem adjungierten Tensor $ \mathrm {adj} (\mathbf {A} ) $:

$ \mathbf {A} ^{-1}={\frac {1}{\det(\mathbf {A} )}}\mathrm {adj} (\mathbf {A} ) $
$ {\begin{aligned}\mathbf {A} =&A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})\\\rightarrow \mathbf {A} ^{-1}=&{\frac {1}{|\mathbf {A} |}}{\begin{pmatrix}A_{22}A_{33}-A_{23}A_{32}&A_{32}A_{13}-A_{33}A_{12}&A_{12}A_{23}-A_{13}A_{22}\\A_{23}A_{31}-A_{21}A_{33}&A_{33}A_{11}-A_{31}A_{13}&A_{13}A_{21}-A_{11}A_{23}\\A_{21}A_{32}-A_{22}A_{31}&A_{31}A_{12}-A_{32}A_{11}&A_{11}A_{22}-A_{12}A_{21}\end{pmatrix}}\end{aligned}} $

Werden die Spalten von A mit Vektoren bezeichnet, also $ \mathbf {A} ={\begin{pmatrix}{\vec {a}}_{1}&{\vec {a}}_{2}&{\vec {a}}_{3}\end{pmatrix}} $, dann gilt:

$ \mathbf {A} ^{-1}={\begin{pmatrix}{\vec {a}}^{1}&{\vec {a}}^{2}&{\vec {a}}^{3}\end{pmatrix}}^{\top }={\frac {1}{\mathrm {det} (\mathbf {A} )}}{\begin{pmatrix}{\vec {a}}_{2}\times {\vec {a}}_{3}&{\vec {a}}_{3}\times {\vec {a}}_{1}&{\vec {a}}_{1}\times {\vec {a}}_{2}\end{pmatrix}}^{\top } $

Satz von Cayley-Hamilton:

$ \mathbf {A} ^{-1}={\frac {1}{\mathrm {I} _{3}(\mathbf {A} )}}(\mathbf {A} ^{2}-\mathrm {I} _{1}(\mathbf {A} )\mathbf {A} +\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} ) $

worin $ \mathrm {I} _{1,2,3} $ die drei #Hauptinvarianten sind.

Inverse des transponierten Tensors:

$ (\mathbf {A} ^{\top })^{-1}=(\mathbf {A} ^{-1})^{\top }=\mathbf {A} ^{\top -1}=\mathbf {A} ^{-\top } $

Inverse eines Tensorprodukts:

$ (\mathbf {A\cdot B} )^{-1}=\mathbf {B} ^{-1}\cdot \mathbf {A} ^{-1} $
$ (x\mathbf {A} )^{-1}={\frac {1}{x}}\mathbf {A} ^{-1} $

#Äußeres Tensorprodukt und Inverse einer Summe:

$ (\mathbf {A+B} )^{-1}={\frac {1}{\det(\mathbf {A+B} )}}\left(\mathrm {adj} (\mathbf {A} )+\mathrm {adj} (\mathbf {B} )+(\mathbf {A} \#\mathbf {B} )^{\top }\right) $

Invertierungsformeln:

$ (a\mathbf {1} +{\vec {b}}\otimes {\vec {c}})^{-1}={\frac {1}{a}}\left(\mathbf {1} -{\frac {1}{a+{\vec {b}}\cdot {\vec {c}}}}{\vec {b}}\otimes {\vec {c}}\right) $
$ {\begin{aligned}&(a\mathbf {1} +{\vec {b}}\otimes {\vec {c}}+{\vec {d}}\otimes {\vec {e}})^{-1}={\frac {1}{aD}}\left(D\mathbf {1} +{\vec {b}}\otimes (q{\vec {c}}+r{\vec {e}})+{\vec {d}}\otimes (s{\vec {c}}+t{\vec {e}})\right)\\&\qquad q=a+{\vec {d}}\cdot {\vec {e}},\quad r=-{\vec {c}}\cdot {\vec {d}},\quad s=-{\vec {b}}\cdot {\vec {e}},\quad t=a+{\vec {b}}\cdot {\vec {c}}\\&\qquad D=rs-qt\end{aligned}} $
$ ({\vec {a}}_{i}\otimes {\vec {g}}_{i})^{-1}={\vec {g}}^{i}\otimes {\vec {a}}^{i} $

Eigensystem

Eigenwertproblem

$ \mathbf {A} \cdot {\hat {v}}=\lambda {\hat {v}} $

mit Eigenwert $ \lambda $ und Eigenvektor $ {\hat {v}} $. Die Eigenvektoren werden auf die Länge eins normiert.

Jeder Tensor hat drei Eigenwerte und drei dazugehörige Eigenvektoren. Mindestens ein Eigenwert und Eigenvektor sind reell. Die beiden anderen Eigenwerte und -vektoren können reell oder komplex sein.

Eigenwerte

Charakteristische Gleichung

$ \mathrm {det} (\mathbf {A} -\lambda _{i}\mathbf {1} )=-\lambda _{i}^{3}+\mathrm {I} _{1}(\mathbf {A} )\lambda _{i}^{2}-\mathrm {I} _{2}(\mathbf {A} )\lambda _{i}+\mathrm {I} _{3}(\mathbf {A} )=0 $

Lösung siehe Cardanische Formeln. Die Koeffizienten sind die #Hauptinvarianten :

$ \mathrm {I} _{1}(\mathbf {A} ):=\mathrm {Sp} (\mathbf {A} )=\lambda _{1}+\lambda _{2}+\lambda _{3} $
$ \mathrm {I} _{2}(\mathbf {A} ):={\frac {1}{2}}[\mathrm {I} _{1}(\mathbf {A} )^{2}-\mathrm {I} _{1}(\mathbf {A} ^{2})]=\lambda _{1}\lambda _{2}+\lambda _{2}\lambda _{3}+\lambda _{3}\lambda _{1} $
$ \mathrm {I} _{3}(\mathbf {A} ):=\mathrm {det} (\mathbf {A} )=\lambda _{1}\lambda _{2}\lambda _{3} $

Eigenvektoren

Eigenvektoren $ {\vec {v}} $ sind nur bis auf einen Faktor ≠ 0 bestimmt. Der Nullvektor ist kein Eigenvektor.

Bestimmungsgleichung: $ (\mathbf {A} -\lambda \mathbf {1} )\cdot {\vec {v}}={\vec {0}} $

Tensor $ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j} $:

$ {\begin{pmatrix}A_{11}-\lambda &A_{12}&A_{13}\\A_{21}&A_{22}-\lambda &A_{23}\\A_{31}&A_{32}&A_{33}-\lambda \end{pmatrix}}\cdot {\begin{pmatrix}v_{1}\\v_{2}\\v_{3}\end{pmatrix}}={\begin{pmatrix}0\\0\\0\end{pmatrix}} $

Bestimmung mit gegebenem/angenommenem $ v_{1} $:

$ {\begin{pmatrix}A_{12}&A_{13}\\A_{22}-\lambda &A_{23}\\A_{32}&A_{33}-\lambda \end{pmatrix}}\cdot {\begin{pmatrix}v_{2}\\v_{3}\end{pmatrix}}=v_{1}{\begin{pmatrix}\lambda -A_{11}\\-A_{21}\\-A_{31}\end{pmatrix}} $

Geometrische Vielfachheit 1:

$ v_{2}=v_{1}{\frac {(\lambda -A_{33})A_{21}+A_{23}A_{31}}{(A_{22}-\lambda )(A_{33}-\lambda )-A_{23}A_{32}}} $
$ v_{3}=v_{1}{\frac {(\lambda -A_{22})A_{31}+A_{32}A_{21}}{(A_{22}-\lambda )(A_{33}-\lambda )-A_{23}A_{32}}} $

Geometrische Vielfachheit 2:

$ {\begin{pmatrix}A_{13}\\A_{23}\\A_{33}-\lambda \end{pmatrix}}v_{3}=-v_{1}{\begin{pmatrix}A_{11}-\lambda \\A_{21}\\A_{31}\end{pmatrix}}-v_{2}{\begin{pmatrix}A_{12}\\A_{22}-\lambda \\A_{32}\end{pmatrix}} $

Die Formeln bleiben richtig, wenn die Indizes {1,2,3} zyklisch vertauscht werden.

Symmetrischen Tensoren: Für das Betragsquadrat der Komponenten $ v_{ij} $ der auf Betrag 1 normierten Eigenvektoren $ {\vec {v}}_{i} $ des (komplexen) Tensors $ A\in \mathbb {C} ^{n\times n} $ gilt mit dessen Eigenwerten $ \lambda _{i} $ und den Eigenwerten $ \mu _{jk} $ der Hauptuntermatrizen von $ A $:[1]

$ |v_{ij}|^{2}\prod _{k=1;k\neq i}^{n}{\big (}\lambda _{i}-\lambda _{k}{\big )}=\prod _{k=1}^{n-1}{\big (}\lambda _{i}-\mu _{jk}{\big )} $

Eigensystem symmetrischer Tensoren

Sei $ \mathbf {A} =\mathbf {A} ^{\top } $ symmetrisch.

Symmetrische Tensoren haben reelle Eigenwerte und paarweise zueinander senkrechte oder orthogonalisierbare Eigenvektoren, die also eine Orthonormalbasis aufbauen. Die Eigenvektoren werden so nummeriert, dass sie ein Rechtssystem bilden.

Hauptachsentransformation mit Eigenwerten $ \lambda _{i} $ und Eigenvektoren $ {\hat {a}}_{i} $ des symmetrischen Tensors A:

$ {\begin{aligned}\mathbf {A} =&\sum _{i=1}^{3}\lambda _{i}{\hat {a}}_{i}\otimes {\hat {a}}_{i}=\left({\hat {a}}_{i}\otimes {\hat {e}}_{i}\right)\cdot \left(\sum _{j=1}^{3}\lambda _{j}{\hat {e}}_{j}\otimes {\hat {e}}_{j}\right)\cdot \left({\hat {e}}_{k}\otimes {\hat {a}}_{k}\right)\\=&{\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}\cdot {\begin{pmatrix}\lambda _{1}&0&0\\0&\lambda _{2}&0\\0&0&\lambda _{3}\end{pmatrix}}\cdot {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}^{\top }\end{aligned}} $

bzw.

$ {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}^{\top }\cdot \mathbf {A} \cdot {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}={\begin{pmatrix}\lambda _{1}&0&0\\0&\lambda _{2}&0\\0&0&\lambda _{3}\end{pmatrix}} $

Eigensystem schiefsymmetrischer Tensoren

Sei $ \mathbf {A} =-\mathbf {A} ^{\top } $ schiefsymmetrisch.

Schiefsymmetrische Tensoren haben einen reellen und zwei konjugiert komplexe, rein imaginäre Eigenwerte. Der reelle Eigenwert von A ist null zu dem ein Eigenvektor gehört, der proportional zur reellen #Vektorinvariante $ {\vec {\mathrm {i} }}(\mathbf {A} ) $ ist. Siehe auch #Axialer Tensor oder Kreuzproduktmatrix.

Eigensystem allgemeiner auch unsymmetrischer Tensoren

Sei $ a,b,c\in \mathbb {R} $ und $ {\vec {a}}_{1},{\vec {a}}_{2},{\vec {a}}_{3}\in \mathbb {R} ^{3} $ eine Basis und $ {\vec {a}}^{1},{\vec {a}}^{2},{\vec {a}}^{3} $ die dazu duale Basis.

Drei reelle Eigenwerte

Der Tensor

$ \mathbf {T} =a\,{\vec {a}}_{1}\otimes {\vec {a}}^{1}+b\,{\vec {a}}_{2}\otimes {\vec {a}}^{2}+c\,{\vec {a}}_{3}\otimes {\vec {a}}^{3} $

hat die Eigenwerte

$ \lambda _{1}=a,\;\lambda _{2}=b,\;\lambda _{3}=c $

und Eigenvektoren

$ {\vec {v}}_{1}={\vec {a}}_{1},\;{\vec {v}}_{2}={\vec {a}}_{2},\;{\vec {v}}_{3}={\vec {a}}_{3} $

Der #transponierte Tensor hat dieselben Eigenwerte zu den dualen Eigenvektoren

$ {\vec {v}}_{1}={\vec {a}}^{1},\;{\vec {v}}_{2}={\vec {a}}^{2},\;{\vec {v}}_{3}={\vec {a}}^{3} $

Ein reeller und zwei konjugiert komplexe Eigenwerte

Der Tensor

$ \mathbf {T} =c\,{\vec {a}}_{1}\otimes {\vec {a}}^{1}+a({\vec {a}}_{2}\otimes {\vec {a}}^{2}+{\vec {a}}_{3}\otimes {\vec {a}}^{3})+b({\vec {a}}_{2}\otimes {\vec {a}}^{3}-{\vec {a}}_{3}\otimes {\vec {a}}^{2}) $

hat die Eigenwerte

$ \lambda _{1}=c,\;\lambda _{2}=a+\mathrm {i} \,b,\;\lambda _{3}=a-\mathrm {i} \,b $

und Eigenvektoren

$ {\vec {v}}_{1}={\vec {a}}_{1},\;{\vec {v}}_{2}={\vec {a}}_{2}+\mathrm {i} \,{\vec {a}}_{3},\;{\vec {v}}_{3}={\vec {a}}_{2}-\mathrm {i} \,{\vec {a}}_{3} $

Der #transponierte Tensor hat dieselben Eigenwerte zu den Eigenvektoren

$ {\vec {v}}_{1}={\vec {a}}^{1},\;{\vec {v}}_{2}={\vec {a}}^{2}-\mathrm {i} \,{\vec {a}}^{3},\;{\vec {v}}_{3}={\vec {a}}^{2}+\mathrm {i} \,{\vec {a}}^{3} $

Invarianten

Eigenwerte des Tensors

Die #Eigenwerte $ \lambda _{1},\lambda _{2},\lambda _{3} $ sind Invarianten.

Hauptinvarianten

#Spur: I1(A), Sp(A)
#Zweite Hauptinvariante: I2(A)
#Determinante: I3(A), det(A), │A

Charakteristisches Polynom

Die Hauptinvarianten des Tensors A sind die Koeffizienten seines charakteristischen Polynoms:

$ \mathrm {det} (\mathbf {A} -x\mathbf {1} )=-x^{3}+\mathrm {Sp} (\mathbf {A} )x^{2}-\mathrm {I} _{2}(\mathbf {A} )x+\mathrm {det} (\mathbf {A} ) $

Spezialfall:

$ \mathrm {det} ({\vec {b}}\otimes {\vec {c}}+a\mathbf {1} )=a^{2}(a+{\vec {b}}\cdot {\vec {c}}) $

Satz von Cayley-Hamilton:

$ -\mathbf {A} ^{3}+\mathrm {Sp} (\mathbf {A} )\mathbf {A} ^{2}-\mathrm {I} _{2}(\mathbf {A} )\mathbf {A} +\mathrm {det} (\mathbf {A} )\mathbf {1} =\mathbf {0} $

Spur

Abbildung $ {\mathcal {L}}\to \mathbb {R} $

$ \mathrm {Sp} (\mathbf {A} )=\mathrm {I} _{1}(\mathbf {A} )={\frac {1}{2}}(\mathbf {A} \#\mathbf {1} ):\mathbf {1} =\lambda _{1}+\lambda _{2}+\lambda _{3} $

mit #Eigenwerten λ1,2,3 von A.

$ \mathrm {Sp} ({\vec {a}}\otimes {\vec {g}})=\mathrm {Sp} ({\vec {g}}\otimes {\vec {a}}):={\vec {a}}\cdot {\vec {g}} $

Linearität: $ x,y\in \mathbb {R} \rightarrow \quad \mathrm {Sp} (x\mathbf {A} +y\mathbf {B} )=x\mathrm {Sp} (\mathbf {A} )+y\mathrm {Sp} (\mathbf {B} ) $

$ \mathrm {Sp} (\mathbf {A} )=\mathrm {Sp} (\mathbf {A} ^{\top }) $
$ \mathrm {Sp} (\mathbf {A\cdot B} )=\mathrm {Sp} (\mathbf {B\cdot A} ) $
$ \mathrm {Sp} (\mathbf {A^{\top }\cdot B} )=\mathrm {Sp} (\mathbf {A\cdot B^{\top }} ) $
$ \mathrm {Sp} (\mathbf {A\cdot B\cdot C} )=\mathrm {Sp} (\mathbf {B\cdot C\cdot A} )=\mathrm {Sp} (\mathbf {C\cdot A\cdot B} ) $

In Komponenten:

$ \mathrm {Sp} \left(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\right)=A_{ii}=A_{11}+A_{22}+A_{33} $
$ \mathrm {Sp} \left(A^{ij}{\vec {a}}_{i}\otimes {\vec {b}}_{j}\right)=A^{ij}{\vec {a}}_{i}\cdot {\vec {b}}_{j} $
$ \mathrm {Sp} \left(A_{j}^{i}{\vec {a}}_{i}\otimes {\vec {a}}^{j}\right)=\mathrm {Sp} \left(A_{i}^{j}{\vec {a}}^{i}\otimes {\vec {a}}_{j}\right)=A_{i}^{i} $

Zweite Hauptinvariante

Abbildung $ {\mathcal {L}}\to \mathbb {R} $

$ \mathrm {I} _{2}(\mathbf {A} ):={\frac {1}{2}}[\mathrm {Sp} (\mathbf {A} )^{2}-\mathrm {Sp} (\mathbf {A} ^{2})]={\frac {1}{2}}(\mathbf {A} \#\mathbf {A} ):\mathbf {1} =\lambda _{1}\lambda _{2}+\lambda _{2}\lambda _{3}+\lambda _{3}\lambda _{1} $

mit #Eigenwerten λ1,2,3 von A.

$ \mathrm {I} _{2}(\mathbf {A} )=\mathrm {Sp(cof} (\mathbf {A} ))=\mathrm {Sp(adj} (\mathbf {A} )) $
$ \mathrm {I} _{2}(x\mathbf {A} )=x^{2}\mathrm {I} _{2}(\mathbf {A} ) $
$ \mathrm {I} _{2}(\mathbf {A} ^{\top })=\mathrm {I} _{2}(\mathbf {A} ) $
$ \mathrm {I} _{2}(\mathbf {A} \cdot \mathbf {B} )=\mathrm {I} _{2}(\mathbf {B} \cdot \mathbf {A} ) $
$ \mathrm {I} _{2}(\mathbf {A} \cdot \mathbf {B} \cdot \mathbf {C} )=\mathrm {I} _{2}(\mathbf {B} \cdot \mathbf {C} \cdot \mathbf {A} )=\mathrm {I} _{2}(\mathbf {C} \cdot \mathbf {A} \cdot \mathbf {B} ) $
$ \mathrm {I} _{2}(\mathbf {A} +\mathbf {B} )=\mathrm {I} _{2}(\mathbf {A} )+\mathrm {I} _{2}(\mathbf {B} )+\mathrm {Sp} (\mathbf {A} )\mathrm {Sp} (\mathbf {B} )-\mathrm {Sp} (\mathbf {A} \cdot \mathbf {B} ) $

In Komponenten:

$ \operatorname {I} _{2}(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{11}A_{22}+A_{11}A_{33}+A_{22}A_{33}-A_{12}A_{21}-A_{13}A_{31}-A_{23}A_{32} $
$ \operatorname {I} _{2}(A^{ij}{\vec {a}}_{i}\otimes {\vec {b}}_{j})={\frac {1}{2}}A^{ij}A^{kl}[({\vec {a}}_{i}\cdot {\vec {b}}_{j})({\vec {a}}_{k}\cdot {\vec {b}}_{l})-({\vec {a}}_{i}\cdot {\vec {b}}_{l})({\vec {a}}_{k}\cdot {\vec {b}}_{j})] $
$ \operatorname {I} _{2}\left(A_{j}^{i}{\vec {a}}_{i}\otimes {\vec {a}}^{j}\right)={\frac {1}{2}}(A_{i}^{i}A_{j}^{j}-A_{j}^{i}A_{i}^{j}) $

Determinante

Abbildung $ {\mathcal {L}}\to \mathbb {R} $

$ \mathrm {I} _{3}(\mathbf {A} ):=\mathrm {det} (\mathbf {A} )={\frac {1}{6}}(\mathbf {A} \#\mathbf {A} ):\mathbf {A} =\lambda _{1}\lambda _{2}\lambda _{3} $

mit #Eigenwerten λ1,2,3 von A.

$ \mathrm {det} (\mathbf {A} ^{\top })=\mathrm {det} (\mathbf {A} ) $

Determinantenproduktsatz:

$ \mathrm {det} (\mathbf {A\cdot B} )=\mathrm {det} (\mathbf {B\cdot A} )=\mathrm {det} (\mathbf {A} )\mathrm {det} (\mathbf {B} ) $
$ \mathrm {det} (\mathbf {A\cdot B\cdot C} )=\mathrm {det} (\mathbf {B\cdot C\cdot A} )=\mathrm {det} (\mathbf {C\cdot A\cdot B} )=\mathrm {det} (\mathbf {A} )\mathrm {det} (\mathbf {B} )\mathrm {det} (\mathbf {C} ) $
$ \mathrm {det} (\mathbf {A} ^{-1})={\frac {1}{\mathrm {det} (\mathbf {A} )}} $

Multiplikation mit Skalaren $ x\in \mathbb {R} $:

$ {\begin{vmatrix}x{\vec {a}}&{\vec {b}}&{\vec {c}}\end{vmatrix}}={\begin{vmatrix}{\vec {a}}&x{\vec {b}}&{\vec {c}}\end{vmatrix}}={\begin{vmatrix}{\vec {a}}&{\vec {b}}&x{\vec {c}}\end{vmatrix}}=x{\begin{vmatrix}{\vec {a}}&{\vec {b}}&{\vec {c}}\end{vmatrix}} $
$ \mathrm {det} (x\mathbf {A} )=x^{3}\mathrm {det} (\mathbf {A} ) $

In Komponenten:

$ {\begin{aligned}\mathrm {det} \left(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\right)=&{\begin{vmatrix}A_{11}&A_{12}&A_{13}\\A_{21}&A_{22}&A_{23}\\A_{31}&A_{32}&A_{33}\end{vmatrix}}\\=&A_{11}(A_{22}A_{33}-A_{23}A_{32})+A_{12}(A_{23}A_{31}-A_{21}A_{33})\\&+A_{13}(A_{21}A_{32}-A_{22}A_{31})\end{aligned}} $
$ {\begin{aligned}\mathrm {det} (A^{ij}{\vec {a}}_{i}\otimes {\vec {g}}_{j})=&{\begin{vmatrix}A^{11}&A^{12}&A^{13}\\A^{21}&A^{22}&A^{23}\\A^{31}&A^{32}&A^{33}\end{vmatrix}}{\begin{vmatrix}{\vec {a}}_{1}&{\vec {a}}_{2}&{\vec {a}}_{3}\end{vmatrix}}{\begin{vmatrix}{\vec {g}}_{1}&{\vec {g}}_{2}&{\vec {g}}_{3}\end{vmatrix}}\end{aligned}} $
$ \operatorname {det} \left(A_{j}^{i}{\vec {a}}_{i}\otimes {\vec {a}}^{j}\right)={\begin{vmatrix}A_{1}^{1}&A_{2}^{1}&A_{3}^{1}\\A_{1}^{2}&A_{2}^{2}&A_{3}^{2}\\A_{1}^{3}&A_{2}^{3}&A_{3}^{3}\end{vmatrix}} $

Zusammenhang mit den anderen Hauptinvarianten:

$ {\begin{aligned}\mathrm {det} (\mathbf {A} )=&{\frac {1}{6}}[\mathrm {Sp} (\mathbf {A} )^{3}-3\mathrm {Sp} (\mathbf {A} )\mathrm {Sp} (\mathbf {A} ^{2})+2\mathrm {Sp} (\mathbf {A} ^{3})]\\[1ex]=&{\frac {1}{3}}[\mathrm {Sp} (\mathbf {A} ^{3})+3\mathrm {Sp} (\mathbf {A} )\mathrm {I} _{2}(\mathbf {A} )-\mathrm {Sp} (\mathbf {A} )^{3}]\end{aligned}} $

Zusammenhang mit dem Spatprodukt:

$ (\mathbf {A} \cdot {\vec {a}})\cdot [(\mathbf {A} \cdot {\vec {b}})\times (\mathbf {A} \cdot {\vec {c}})]=\mathrm {det} (\mathbf {A} ){\vec {a}}\cdot ({\vec {b}}\times {\vec {c}}) $

Zusammenhang mit #Äußeres Tensorprodukt:

$ \det(\mathbf {A} )={\frac {1}{6}}(\mathbf {A} \#\mathbf {A} ):\mathbf {A} $
$ {\begin{aligned}\rightarrow \det(\mathbf {A+B} )=&\det(\mathbf {A} )+\det(\mathbf {B} )+\mathrm {Sp} (\mathbf {A} )\mathrm {I} _{2}(\mathbf {B} )+\mathrm {I} _{2}(\mathbf {A} )\mathrm {Sp} (\mathbf {B} )\\&+\mathrm {Sp} (\mathbf {A\cdot B\cdot (A+B)} )-\mathrm {Sp} (\mathbf {A\cdot B} )\mathrm {Sp} (\mathbf {A+B} )\end{aligned}} $

Zusammenhang mit dem #Kofaktor:

$ \det(\mathbf {A} +\mathbf {B} )=\det(\mathbf {A} )+\mathrm {cof} (\mathbf {A} ):\mathbf {B} +\mathbf {A} :\mathrm {cof} (\mathbf {B} )+\det(\mathbf {B} ) $

Betrag

Abbildung $ {\mathcal {L}}\to \mathbb {R} $

$ \parallel \mathbf {A} \parallel :={\sqrt {\mathbf {A} :\mathbf {A} }}={\sqrt {\mathrm {Sp} (\mathbf {A} ^{\top }\cdot \mathbf {A} )}} $
$ \parallel {\vec {a}}\otimes {\vec {g}}\parallel =|{\vec {a}}|\,|{\vec {g}}| $
$ \parallel A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\parallel ={\sqrt {A_{ij}A_{ij}}} $
$ \parallel A^{ij}{\vec {a}}_{i}\otimes {\vec {g}}_{j}\parallel ={\sqrt {A^{ij}A^{kl}({\vec {a}}_{i}\cdot {\vec {a}}_{k})({\vec {g}}_{j}\cdot {\vec {g}}_{l})}} $
$ \parallel A_{j}^{i}{\vec {a}}_{i}\otimes {\vec {a}}^{j}\parallel ={\sqrt {A_{j}^{i}A_{l}^{k}({\vec {a}}_{i}\cdot {\vec {a}}_{k})({\vec {a}}^{j}\cdot {\vec {a}}^{l})}} $

Falls $ \mathbf {A} =\mathbf {A} ^{\top } $:

$ \quad \parallel \mathbf {A} \parallel ={\sqrt {\mathrm {Sp} ^{2}(\mathbf {A} )-2\mathrm {I} _{2}(\mathbf {A} )}}={\sqrt {\mathrm {Sp} (\mathbf {A} ^{2})}}={\sqrt {\lambda _{1}^{2}+\lambda _{2}^{2}+\lambda _{3}^{2}}} $

Falls $ \mathbf {A} =-\mathbf {A} ^{\top } $:

$ \quad \parallel \mathbf {A} \parallel ={\sqrt {2\mathrm {I} _{2}(\mathbf {A} )}}={\sqrt {-\mathrm {Sp} (\mathbf {A} ^{2})}} $

Dualer axialer Vektor

Für #Schiefsymmetrische Tensoren $ \mathbf {A} =-\mathbf {A} ^{\top }=\mathbf {A} ^{\mathrm {A} } $ gibt es einen dualen axialen Vektor $ {\stackrel {A}{\overrightarrow {\mathbf {A} }}} $ für den gilt:

$ \mathbf {A} \cdot {\vec {v}}={\stackrel {A}{\overrightarrow {\mathbf {A} }}}\times {\vec {v}} $ für alle $ {\vec {v}}\in \mathbb {V} $

Der duale axiale Vektor ist proportional zur #Vektorinvariante:

$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}}:=-{\frac {1}{2}}{\vec {\mathrm {i} }}(\mathbf {A} ) $

Berechnung mit #Fundamentaltensor 3. Stufe $ {\stackrel {3}{\mathbf {E} }} $, #Kreuzprodukt von Tensoren oder #Skalarkreuzprodukt von Tensoren:

$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}}=-{\frac {1}{2}}{\stackrel {3}{\mathbf {E} }}:\mathbf {A} =-{\frac {1}{2}}\mathbf {A} \times \mathbf {1} =-{\frac {1}{2}}\mathbf {A} \cdot \!\!\times \mathbf {1} $
$ {\stackrel {A}{\overrightarrow {A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}}}}=-{\frac {1}{2}}A_{ij}{\hat {e}}_{i}\times {\hat {e}}_{j}={\frac {1}{2}}{\begin{pmatrix}A_{32}-A_{23}\\A_{13}-A_{31}\\A_{21}-A_{12}\end{pmatrix}} $
$ {\stackrel {A}{\overrightarrow {A^{ij}({\vec {a}}_{i}\otimes {\vec {b}}_{j})}}}=-{\frac {1}{2}}A^{ij}{\vec {a}}_{i}\times {\vec {b}}_{j} $

#Symmetrische Tensoren und #Kugeltensoren haben keinen dualen axialen Vektor: $ {\stackrel {A}{\overrightarrow {\mathbf {A} ^{\mathrm {S} }}}}={\stackrel {A}{\overrightarrow {\mathbf {A} ^{\mathrm {K} }}}}={\vec {0}} $

Ein #Symmetrischer Anteil oder #Kugelanteil trägt nichts zum dualen axialen Vektor bei: $ {\stackrel {A}{\overrightarrow {\mathbf {A} }}}={\stackrel {A}{\overrightarrow {\mathbf {A} ^{\mathrm {D} }}}}={\stackrel {A}{\overrightarrow {\mathbf {A} ^{\mathrm {A} }}}} $

Seien x eine beliebige Zahl, $ {\vec {u}},\,{\vec {v}} $ beliebige Vektoren und A, B beliebige Tensoren zweiter Stufe. Dann gilt:

$ {\stackrel {A}{\overrightarrow {{\vec {u}}\otimes {\vec {v}}}}}\;={\frac {1}{2}}{\vec {v}}\times {\vec {u}} $
$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}}\times {\vec {v}}=\mathbf {A} ^{\mathrm {A} }\cdot {\vec {v}} $
$ {\stackrel {A}{\overrightarrow {\mathbf {A} ^{\top }}}}\quad \;=-{\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
$ {\stackrel {A}{\overrightarrow {\mathbf {A+B} }}}={\stackrel {A}{\overrightarrow {\mathbf {A} }}}+{\stackrel {A}{\overrightarrow {\mathbf {B} }}} $
$ {\stackrel {A}{\overrightarrow {x\mathbf {A} }}}\quad \;=x\,{\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
$ {\stackrel {A}{\overrightarrow {\mathbf {A} \#\mathbf {B} }}}\;=\mathbf {A} \cdot {\stackrel {A}{\overrightarrow {\mathbf {B} }}}+\mathbf {B} \cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
$ \mathbf {A} \cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}}\;=\mathbf {A} ^{\top }\cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}}={\stackrel {A}{\overrightarrow {\mathbf {A} }}}\cdot \mathbf {A} $
$ {\stackrel {A}{\overrightarrow {\mathrm {cof} (\mathbf {A} )}}}=\mathbf {A} \cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
$ {\stackrel {A}{\overrightarrow {\mathbf {A} ^{-1}}}}\quad =-{\frac {1}{\mathrm {det} (\mathbf {A} )}}\mathbf {A} \cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}}\quad {\text{falls}}\quad \mathrm {det} (\mathbf {A} )\neq 0 $
$ {\stackrel {A}{\overrightarrow {{\vec {v}}\times \mathbf {A} }}}={\frac {1}{2}}(\mathrm {Sp} (\mathbf {A} )\mathbf {1} -\mathbf {A} )\cdot {\vec {v}}={\frac {1}{2}}(\mathbf {A} ^{\top }\#\mathbf {1} )\cdot {\vec {v}} $
$ {\stackrel {A}{\overrightarrow {{\vec {v}}\times \mathbf {1} }}}\;\;={\vec {v}} $
$ {\stackrel {A}{\overrightarrow {({\vec {u}}\times {\vec {v}})\times \mathbf {A} }}}={\frac {1}{2}}({\vec {u}}\cdot \mathbf {A} \times {\vec {v}}-{\vec {v}}\cdot \mathbf {A} \times {\vec {u}}) $
$ {\stackrel {A}{\overrightarrow {\mathbf {B\cdot A\cdot B} ^{\top }}}}\;\;=\mathrm {cof} (\mathbf {B} )\cdot {\stackrel {A}{\overrightarrow {\mathbf {A} }}} $

Darin ist „#“ ein #Äußeres Tensorprodukt, cof(·) ist der #Kofaktor.

Vektorinvariante

$ {\vec {\mathrm {i} }}(\mathbf {A} ):=\mathbf {A} \cdot \!\!\times \mathbf {1} =\mathbf {A} \times \mathbf {1} ={\stackrel {3}{\mathbf {E} }}:\mathbf {A} =-2{\stackrel {A}{\overrightarrow {\mathbf {A} }}} $
$ {\vec {\mathrm {i} }}(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{ij}{\hat {e}}_{i}\times {\hat {e}}_{j}={\begin{pmatrix}A_{23}-A_{32}\\A_{31}-A_{13}\\A_{12}-A_{21}\end{pmatrix}} $
$ {\vec {\mathrm {i} }}(A^{ij}({\vec {a}}_{i}\otimes {\vec {b}}_{j}))=A^{ij}{\vec {a}}_{i}\times {\vec {b}}_{j} $

Zusammenhang mit dem #Skalarkreuzprodukt von Tensoren:

$ \mathbf {A} \cdot \!\!\times \mathbf {B} ={\vec {\mathrm {i} }}(\mathbf {A} \cdot \mathbf {B} ) $

#Symmetrische Tensoren haben keine Vektorinvariante: $ {\vec {\mathrm {i} }}(\mathbf {A} ^{\mathrm {S} })={\vec {0}} $

Die Eigenschaften des dualen axialen Vektors sind hierher übertragbar. Seien x eine beliebige Zahl, $ {\vec {u}},\,{\vec {v}} $ beliebige Vektoren und A, B beliebige Tensoren zweiter Stufe. Dann gilt:

$ {\vec {\mathrm {i} }}({\vec {u}}\otimes {\vec {v}})\;\;={\vec {u}}\times {\vec {v}} $
$ {\vec {\mathrm {i} }}(\mathbf {A} )\times {\vec {v}}\;=-2\mathbf {A} ^{\mathrm {A} }\cdot {\vec {v}}=(\mathbf {A^{\top }-A} )\cdot {\vec {v}} $
$ {\vec {\mathrm {i} }}(\mathbf {A} ^{\top })\quad \;=-{\vec {\mathrm {i} }}(\mathbf {A} ) $
$ {\vec {\mathrm {i} }}(\mathbf {A+B} )={\vec {\mathrm {i} }}(\mathbf {A} )+{\vec {\mathrm {i} }}(\mathbf {B} ) $
$ {\vec {\mathrm {i} }}(x\mathbf {A} )\quad \;=x\,{\vec {\mathrm {i} }}(\mathbf {A} ) $
$ {\vec {\mathrm {i} }}(\mathbf {A} \#\mathbf {B} )\;=\mathbf {A} \cdot {\vec {\mathrm {i} }}(\mathbf {B} )+\mathbf {B} \cdot {\vec {\mathrm {i} }}(\mathbf {A} ) $
$ \mathbf {A} \cdot {\vec {\mathrm {i} }}(\mathbf {A} )\;=\mathbf {A} ^{\top }\cdot {\vec {\mathrm {i} }}(\mathbf {A} )={\vec {\mathrm {i} }}(\mathbf {A} )\cdot \mathbf {A} $
$ {\vec {\mathrm {i} }}(\mathrm {cof} (\mathbf {A} ))=\mathbf {A} \cdot {\vec {\mathrm {i} }}(\mathbf {A} ) $
$ {\vec {\mathrm {i} }}(\mathbf {A} ^{-1})\quad =-{\frac {1}{\mathrm {det} (\mathbf {A} )}}\mathbf {A} \cdot {\vec {\mathrm {i} }}(\mathbf {A} )\quad {\text{falls}}\quad \mathrm {det} (\mathbf {A} )\neq 0 $
$ {\vec {\mathrm {i} }}({\vec {v}}\times \mathbf {A} )\;=(\mathbf {A} -\mathrm {Sp} (\mathbf {A} )\mathbf {1} )\cdot {\vec {v}}=-(\mathbf {A} ^{\top }\#\mathbf {1} )\cdot {\vec {v}} $
$ {\vec {\mathrm {i} }}({\vec {v}}\times \mathbf {1} )\;\;=-2{\vec {v}} $
$ {\vec {\mathrm {i} }}(({\vec {u}}\times {\vec {v}})\times \mathbf {A} )={\vec {v}}\cdot \mathbf {A} \times {\vec {u}}-{\vec {u}}\cdot \mathbf {A} \times {\vec {v}} $
$ {\vec {\mathrm {i} }}(\mathbf {B\cdot A\cdot B} ^{\top })=\mathrm {cof} (\mathbf {B} )\cdot {\vec {\mathrm {i} }}(\mathbf {A} ) $

Darin ist „#“ ein #Äußeres Tensorprodukt, cof(·) ist der #Kofaktor.

Spezielle Tensoren

Dyade

Definition

$ \mathbf {A} :={\vec {a}}\otimes {\vec {b}} $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathbf {0} $

#Invarianten:

$ \mathrm {Sp} (\mathbf {A} )={\vec {a}}\cdot {\vec {b}} $
$ \mathrm {I} _{2}(\mathbf {A} )=0 $
$ \mathrm {det} (\mathbf {A} )=0 $
$ \parallel \mathbf {A} \parallel =|{\vec {a}}|\,|{\vec {b}}| $

#Eigensystem:

$ {\begin{aligned}\lambda _{1}=&{\vec {a}}\cdot {\vec {b}},&{\vec {v}}_{1}=&{\frac {\vec {a}}{|{\vec {a}}|}}\\\lambda _{2}=&0,&{\vec {v}}_{2}=&{\frac {{\vec {a}}\times {\vec {b}}}{|{\vec {a}}\times {\vec {b}}|}}\\\lambda _{3}=&0,&{\vec {v}}_{3}=&{\frac {({\vec {a}}\times {\vec {b}})\times {\vec {b}}}{|({\vec {a}}\times {\vec {b}})\times {\vec {b}}|}}\end{aligned}} $

Dyadentripel

Gegeben ein beliebiger Tensor 2. Stufe A. Dieser kann immer als Summe dreier Dyaden dargestellt werden:

$ {\begin{aligned}\mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}=&{\vec {s}}_{j}\otimes {\hat {e}}_{j}={\begin{pmatrix}{\vec {s}}_{1}&{\vec {s}}_{2}&{\vec {s}}_{3}\end{pmatrix}}\\=&{\hat {e}}_{i}\otimes {\vec {z}}_{i}={\begin{pmatrix}{\vec {z}}_{1}&{\vec {z}}_{2}&{\vec {z}}_{3}\end{pmatrix}}^{\top }\\=&{\vec {a}}_{k}\otimes {\vec {g}}_{k}\end{aligned}} $

mit Spaltenvektoren $ {\vec {s}}_{j}=A_{ij}{\hat {e}}_{i}=\mathbf {A} \cdot {\hat {e}}_{j} $, Zeilenvektoren $ {\vec {z}}_{i}=A_{ij}{\hat {e}}_{j}={\hat {e}}_{i}\cdot \mathbf {A} $ und $ {\vec {g}}_{k}=({\vec {a}}^{k}\cdot {\hat {e}}_{i})A_{ij}{\hat {e}}_{j}={\vec {a}}^{k}\cdot \mathbf {A} $.

#Hauptinvarianten ($ x_{m,n}:={\vec {x}}_{m}\cdot {\hat {e}}_{n} $):

$ \mathrm {I} _{1}(\mathbf {A} )=s_{i,i}=z_{i,i}={\vec {a}}_{i}\cdot {\vec {g}}_{i} $
$ {\begin{aligned}\mathrm {I} _{2}(\mathbf {A} )=&{\frac {1}{2}}(s_{i,i}s_{j,j}-s_{i,j}s_{j,i})={\frac {1}{2}}(z_{i,i}z_{j,j}-z_{i,j}z_{j,i})\\=&{\frac {1}{2}}[({\vec {a}}_{i}\cdot {\vec {g}}_{i})({\vec {a}}_{j}\cdot {\vec {g}}_{j})-({\vec {a}}_{i}\cdot {\vec {g}}_{j})({\vec {a}}_{j}\cdot {\vec {g}}_{i})]\end{aligned}} $
$ \mathrm {I} _{3}(\mathbf {A} )={\begin{vmatrix}{\vec {s}}_{1}&{\vec {s}}_{2}&{\vec {s}}_{3}\end{vmatrix}}={\begin{vmatrix}{\vec {z}}_{1}&{\vec {z}}_{2}&{\vec {z}}_{3}\end{vmatrix}}={\begin{vmatrix}{\vec {a}}_{1}&{\vec {a}}_{2}&{\vec {a}}_{3}\end{vmatrix}}{\begin{vmatrix}{\vec {g}}_{1}&{\vec {g}}_{2}&{\vec {g}}_{3}\end{vmatrix}} $

#Betrag:

$ \parallel \mathbf {A} \parallel ={\sqrt {{\vec {s}}_{i}\cdot {\vec {s}}_{i}}}={\sqrt {{\vec {z}}_{i}\cdot {\vec {z}}_{i}}}={\sqrt {({\vec {a}}_{i}\cdot {\vec {a}}_{j})({\vec {g}}_{i}\cdot {\vec {g}}_{j})}} $

#Dualer axialer Vektor:

$ {\stackrel {A}{\overrightarrow {\mathbf {A} }}}={\frac {1}{2}}{\hat {e}}_{i}\times {\vec {s}}_{i}={\frac {1}{2}}{\vec {z}}_{i}\times {\hat {e}}_{i}={\frac {1}{2}}{\vec {g}}_{i}\times {\vec {a}}_{i} $

#Vektorinvariante:

$ {\vec {\mathrm {i} }}(\mathbf {A} )={\vec {s}}_{i}\times {\hat {e}}_{i}={\hat {e}}_{i}\times {\vec {z}}_{i}={\vec {a}}_{i}\times {\vec {g}}_{i} $

#Kofaktor:

$ {\begin{aligned}\mathrm {cof} (\mathbf {A} )=&{\vec {z}}_{i}\otimes {\vec {s}}_{i}-\mathrm {I} _{1}(\mathbf {A} )\mathbf {A} ^{\top }+\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} \\=&({\vec {a}}_{i}\cdot {\vec {g}}_{j}){\vec {g}}_{i}\otimes {\vec {a}}_{j}-({\vec {a}}_{i}\cdot {\vec {g}}_{i}){\vec {g}}_{j}\otimes {\vec {a}}_{j}+\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} \end{aligned}} $

#Inverse:

$ \mathbf {A} ^{-1}={\hat {e}}_{i}\otimes {\vec {s}}^{i}={\vec {z}}^{i}\otimes {\hat {e}}_{i}={\vec {g}}^{i}\otimes {\vec {a}}^{i} $

Einheitstensor

$ \mathbf {1} ={\hat {e}}_{i}\otimes {\hat {e}}_{i}=\delta _{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}={\begin{pmatrix}1&0&0\\0&1&0\\0&0&1\end{pmatrix}} $
$ \mathbf {1} ={\vec {g}}_{i}\otimes {\vec {g}}^{i}={\vec {g}}^{i}\otimes {\vec {g}}_{i}=g^{ij}{\vec {g}}_{i}\otimes {\vec {g}}_{j}=g_{ij}{\vec {g}}^{i}\otimes {\vec {g}}^{j} $

mit $ g_{ij}={\vec {g}}_{i}\cdot {\vec {g}}_{j}\,,\;g^{ij}={\vec {g}}^{i}\cdot {\vec {g}}^{j} $

Allgemein:

$ \mathbf {1} =({\vec {a}}^{i}\cdot {\vec {g}}^{j}){\vec {a}}_{i}\otimes {\vec {g}}_{j} $

#Transposition und #Inverse:

$ \mathbf {1} =\mathbf {1} ^{\top }=\mathbf {1} ^{-1}=\mathbf {1} ^{\top -1} $

Kofaktor: $ \mathrm {cof} (\mathbf {1} )=\mathbf {1} $

Vektortransformation

$ \mathbf {1} \cdot {\vec {v}}={\vec {v}}\cdot \mathbf {1} ={\vec {v}} $

Tensorprodukt

$ \mathbf {A\cdot 1} =\mathbf {1\cdot A} =\mathbf {A} $

Skalarprodukt

$ \mathbf {A} :\mathbf {1} =\mathrm {Sp} (\mathbf {A} ) $

#Invarianten:

$ \mathrm {Sp} (\mathbf {1} )=\mathbf {1} :\mathbf {1} =3 $
$ \mathrm {I} _{2}(\mathbf {1} )=3 $
$ \mathrm {det} (\mathbf {1} )=1 $
$ \parallel \mathbf {1} \parallel ={\sqrt {3}} $

#Eigenwerte:

$ \lambda _{1,2,3}=1 $

Alle Vektoren sind #Eigenvektoren.

Unimodulare Tensoren

Definition

$ \mathbf {H} :\quad \mathrm {det} (\mathbf {H} )=1 $

Kofaktor: $ \mathrm {cof} (\mathbf {H} )=\mathbf {H} ^{\top -1} $

Determinantenproduktsatz:

$ \mathrm {det} (\mathbf {A\cdot H} )=\mathrm {det} (\mathbf {H\cdot A} )=\mathrm {det} (\mathbf {A} ) $

Orthogonale Tensoren

Definition

$ \mathbf {Q} :\quad \mathbf {Q} ^{-1}=\mathbf {Q} ^{\top }\quad {\textsf {oder}}\quad \mathbf {Q\cdot Q} ^{\top }=\mathbf {Q} ^{\top }\cdot \mathbf {Q} =\mathbf {1} $

Kofaktor: $ \mathrm {cof} (\mathbf {Q} )=\mathrm {det} (\mathbf {Q} )\mathbf {Q} =\pm \mathbf {Q} $

#Invarianten ($ \alpha $ ist der Drehwinkel):

$ \mathrm {Sp} (\mathbf {Q} )=\mathrm {det} (\mathbf {Q} )+2\cos(\alpha ) $
$ \mathrm {I} _{2}(\mathbf {Q} )=\mathrm {det} (\mathbf {Q} )\cdot \mathrm {Sp} (\mathbf {Q} )=1+2\,\mathrm {det} (\mathbf {Q} )\cos(\alpha ) $
$ \mathrm {det} (\mathbf {Q} )=\pm 1 $
$ \parallel \mathbf {Q} \parallel ={\sqrt {3}} $

Eigentlich orthogonaler Tensor $ \mathrm {det} (\mathbf {Q} )=+1 $, entspricht einer Drehung.

Uneigentlich orthogonaler Tensor $ \mathrm {det} (\mathbf {Q} )=-1 $, entspricht einer Drehspiegelung.

Spatprodukt:

$ (\mathbf {Q} \cdot {\vec {a}})\cdot [(\mathbf {Q} \cdot {\vec {b}})\times (\mathbf {Q} \cdot {\vec {c}})]=\mathrm {det} (\mathbf {Q} ){\vec {a}}\cdot ({\vec {b}}\times {\vec {c}}) $

Kreuzprodukt und #Kofaktor:

$ (\mathbf {Q} \cdot {\vec {a}})\times (\mathbf {Q} \cdot {\vec {b}})=\mathrm {det} (\mathbf {Q} )\mathbf {Q} \cdot ({\vec {a}}\times {\vec {b}}) $
$ \mathrm {cof} (\mathbf {Q} )=\mathrm {det} (\mathbf {Q} )\mathbf {Q} $

Gegeben ein Einheitsvektor $ {\hat {n}}={\begin{pmatrix}n_{1}&n_{2}&n_{3}\end{pmatrix}}^{\top } $ und Drehwinkel α. Dann sind die folgenden Tensoren R zueinander gleich, orthogonal und drehen um die Achse $ {\hat {n}} $ mit Winkel α:

Rodrigues-Formel:

$ {\begin{aligned}\mathbf {R} =&\mathbf {1} +s_{\alpha }{\hat {n}}\times \mathbf {1} +d_{\alpha }({\hat {n}}\times \mathbf {1} )^{2}\\=&\mathbf {1} +s_{\alpha }{\hat {n}}\times \mathbf {1} +d_{\alpha }({\hat {n}}\otimes {\hat {n}}-\mathbf {1} )\end{aligned}} $
$ \mathbf {R} ={\begin{pmatrix}c_{\alpha }+d_{\alpha }n_{1}^{2}&-s_{\alpha }n_{3}+d_{\alpha }n_{1}n_{2}&s_{\alpha }n_{2}+d_{\alpha }n_{1}n_{3}\\s_{\alpha }n_{3}+d_{\alpha }n_{1}n_{2}&c_{\alpha }+d_{\alpha }n_{2}^{2}&-s_{\alpha }n_{1}+d_{\alpha }n_{2}n_{3}\\-s_{\alpha }n_{2}+d_{\alpha }n_{1}n_{3}&s_{\alpha }n_{1}+d_{\alpha }n_{2}n_{3}&c_{\alpha }+d_{\alpha }n_{3}^{2}\end{pmatrix}} $

mit $ c_{\alpha }=\cos(\alpha ),\;d_{\alpha }=1-\cos(\alpha ),\;s_{\alpha }=\sin(\alpha ) $.

Euler-Rodrigues-Formel: $ a=\cos \left({\tfrac {\alpha }{2}}\right),b=\sin \left({\tfrac {\alpha }{2}}\right)n_{1},c=\sin \left({\tfrac {\alpha }{2}}\right)n_{2},d=\sin \left({\tfrac {\alpha }{2}}\right)n_{3} $ also $ a^{2}+b^{2}+c^{2}+d^{2}=1 $:

$ \mathbf {R} :={\begin{pmatrix}a^{2}+b^{2}-c^{2}-d^{2}&2(bc-ad)&2(bd+ac)\\2(bc+ad)&a^{2}+c^{2}-b^{2}-d^{2}&2(cd-ab)\\2(bd-ac)&2(cd+ab)&a^{2}+d^{2}-b^{2}-c^{2}\end{pmatrix}} $

Formulierung mit Drehvektor:

Drehvektor Orthogonaler Tensor
$ {\vec {\alpha }}=\alpha {\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +{\frac {\sin(\alpha )}{\alpha }}{\vec {\alpha }}\times \mathbf {1} +{\frac {1-\cos(\alpha )}{\alpha ^{2}}}({\vec {\alpha }}\times \mathbf {1} )^{2} $
$ {\vec {\alpha }}=\tan(\alpha ){\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +\cos(\alpha ){\vec {\alpha }}\times \mathbf {1} +{\frac {\cos ^{2}(\alpha )}{1+\cos(\alpha )}}({\vec {\alpha }}\times \mathbf {1} )^{2} $
$ {\vec {\alpha }}=\tan \left({\frac {\alpha }{2}}\right)\;{\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +{\frac {2}{1+{\vec {\alpha }}\cdot {\vec {\alpha }}}}({\vec {\alpha }}\times \mathbf {1} +({\vec {\alpha }}\times \mathbf {1} )^{2}) $
$ {\vec {\alpha }}=\sin(\alpha )\;{\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +{\vec {\alpha }}\times \mathbf {1} +{\dfrac {1}{1+\cos(\alpha )}}({\vec {\alpha }}\times \mathbf {1} )^{2} $
$ {\vec {\alpha }}=\sin \left({\frac {\alpha }{2}}\right)\;{\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +2\cos \left({\frac {\alpha }{2}}\right){\vec {\alpha }}\times \mathbf {1} +2({\vec {\alpha }}\times \mathbf {1} )^{2} $
$ {\vec {\alpha }}=\cos(\alpha )\;{\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +\tan(\alpha ){\vec {\alpha }}\times \mathbf {1} +{\frac {1-\cos(\alpha )}{{\vec {\alpha }}\cdot {\vec {\alpha }}}}({\vec {\alpha }}\times \mathbf {1} )^{2} $
$ {\vec {\alpha }}=\cos \left({\frac {\alpha }{2}}\right)\;{\vec {n}} $  →  $ \mathbf {R} =\mathbf {1} +2\sin \left({\frac {\alpha }{2}}\right){\vec {\alpha }}\times \mathbf {1} +2{\frac {1-{\vec {\alpha }}\cdot {\vec {\alpha }}}{{\vec {\alpha }}\cdot {\vec {\alpha }}}}({\vec {\alpha }}\times \mathbf {1} )^{2} $

Darin ist $ ({\vec {\alpha }}\times \mathbf {1} )^{2}=({\vec {\alpha }}\times \mathbf {1} )\cdot ({\vec {\alpha }}\times \mathbf {1} )={\vec {\alpha }}\otimes {\vec {\alpha }}-({\vec {\alpha }}\cdot {\vec {\alpha }})\mathbf {1} $

Beispiel für Drehspiegelung:

$ \mathbf {Q} =-\mathbf {1} +\sin(\alpha ){\hat {n}}\times \mathbf {1} -(1+\cos(\alpha ))({\hat {n}}\times \mathbf {1} )^{2} $

Drehung von Vektorraumbasis $ {\vec {u}}_{1,2,3}\;{\textsf {nach}}\;{\vec {v}}_{1,2,3} $ mit Drehachse $ {\hat {n}} $:

$ \mathbf {Q} \cdot {\vec {u}}_{i}={\vec {v}}_{i}\,,\quad \mathbf {Q} \cdot {\vec {u}}^{i}={\vec {v}}^{i}\,,\quad \mathbf {Q} ={\vec {v}}_{i}\otimes {\vec {u}}^{i}={\vec {v}}^{i}\otimes {\vec {u}}_{i} $
$ {\hat {n}}\simeq {\vec {v}}_{i}\times {\vec {u}}^{i}={\vec {v}}^{i}\times {\vec {u}}_{i}=-2{\stackrel {A}{\overrightarrow {\mathbf {Q} }}}={\vec {\mathrm {i} }}(\mathbf {Q} ) $

mit #Dualer axialer Vektor $ {\stackrel {A}{\overrightarrow {\mathbf {Q} }}} $ und #Vektorinvariante $ {\vec {\mathrm {i} }}(\mathbf {Q} ) $.

Gegeben Orthonormalbasis $ {\hat {v}}_{1,2,3} $, Drehwinkel $ \alpha $ und $ {\hat {v}}_{1} $ ist Drehachse:

$ {\begin{aligned}\mathbf {Q} =&{\color {red}\pm }{\hat {v}}_{1}\otimes {\hat {v}}_{1}+\cos(\alpha )({\hat {v}}_{2}\otimes {\hat {v}}_{2}+{\hat {v}}_{3}\otimes {\hat {v}}_{3})+\sin(\alpha )({\hat {v}}_{3}\otimes {\hat {v}}_{2}-{\hat {v}}_{2}\otimes {\hat {v}}_{3})\\=&{\begin{pmatrix}{\color {red}\pm 1}&0&0\\0&\cos(\alpha )&-\sin(\alpha )\\0&\sin(\alpha )&\cos(\alpha )\end{pmatrix}}_{{\hat {v}}_{i}\otimes {\hat {v}}_{j}}\end{aligned}} $
$ {\color {red}+1} $: Drehung, $ {\color {red}-1} $: Drehspiegelung um $ {\hat {v}}_{1} $

Wenn $ {\hat {v}}_{1,2,3} $ ein Rechtssystem (Mathematik) bilden, dann dreht Q gegen den Uhrzeigersinn, sonst im Uhrzeigersinn um die Drehachse.

#Eigensystem:

$ {\begin{aligned}\lambda _{1}=&\mathrm {det} (\mathbf {Q} )\,,&{\vec {q}}_{1}=&{\hat {v}}_{1}\\\lambda _{2}=&e^{\mathrm {i} \alpha },&{\vec {q}}_{2}=&{\frac {1}{\sqrt {2}}}({\hat {v}}_{2}-\mathrm {i} {\hat {v}}_{3}).\\\lambda _{3}=&e^{-\mathrm {i} \alpha },&{\vec {q}}_{3}=&{\frac {1}{\sqrt {2}}}({\hat {v}}_{2}+\mathrm {i} {\hat {v}}_{3})\end{aligned}} $

Drehwinkel:

$ \cos(\alpha )={\frac {1}{2}}(\mathrm {Sp} (\mathbf {Q} )-\mathrm {det} (\mathbf {Q} )) $

Drehachse $ {\hat {n}} $ ist #Vektorinvariante:

$ {\hat {n}}\simeq {\vec {\mathrm {i} }}(\mathbf {Q} )=\mathbf {1} \cdot \!\!\times \mathbf {Q} $
$ \mathbf {Q} ={\vec {s}}_{i}\otimes {\vec {e}}_{i}={\vec {e}}_{i}\otimes {\vec {z}}_{i}\quad \rightarrow \quad {\hat {n}}\simeq {\vec {s}}_{i}\times {\vec {e}}_{i}={\vec {e}}_{i}\times {\vec {z}}_{i} $
$ {\frac {1}{2}}(\mathbf {Q} -\mathbf {Q} ^{\top })=\sin(\alpha ){\hat {n}}\times \mathbf {1} =\sin(\alpha ){\begin{pmatrix}0&-n_{3}&n_{2}\\n_{3}&0&-n_{1}\\-n_{2}&n_{1}&0\end{pmatrix}},\quad |{\hat {n}}|=1 $

Positiv definite Tensoren

Definition

$ \mathbf {A} :\quad {\vec {v}}\cdot \mathbf {A} \cdot {\vec {v}}>0\quad \forall \;{\vec {v}}\in \mathbb {V} \setminus \{{\vec {0}}\} $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathbf {A^{\top }\cdot A^{\top }} -\mathrm {I} _{1}(\mathbf {A} )\mathbf {A} ^{\top }+\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} =\det(\mathbf {A} )\mathbf {A} ^{\top -1} $

Notwendige Bedingungen für positive Definitheit:

$ \mathrm {det} (\mathbf {A} )>0 $
$ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\quad \rightarrow \quad A_{11},\,A_{22},\,A_{33}>0 $
$ \mathbf {A} =A_{j}^{i}{\vec {a}}_{i}\otimes {\vec {a}}^{j}\quad \rightarrow \quad A_{1}^{1},\,A_{2}^{2},\,A_{3}^{3}>0 $

Notwendige und hinreichende Bedingung für positive Definitheit: Alle #Eigenwerte von A sind größer als null.

Immer positiv definit falls det(A) ≠ 0:

A·A und A·A

Symmetrische Tensoren

Definition

$ \mathbf {A} :\quad \mathbf {A} =\mathbf {A} ^{\top } $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathrm {adj} (\mathbf {A} )=\mathbf {A} ^{2}-\mathrm {Sp} (\mathbf {A} )\mathbf {A} +\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} $

#Betrag:

$ \quad \parallel \mathbf {A} \parallel ={\sqrt {\mathrm {Sp} ^{2}(\mathbf {A} )-2\mathrm {I} _{2}(\mathbf {A} )}}={\sqrt {\mathrm {Sp} (\mathbf {A} ^{2})}}={\sqrt {\lambda _{1}^{2}+\lambda _{2}^{2}+\lambda _{3}^{2}}} $

Bei Symmetrischen Tensoren verschwinden ihr #Dualer axialer Vektor und ihre #Vektorinvariante:

$ {\stackrel {A}{\overrightarrow {\mathbf {A} ^{\mathrm {S} }}}}={\vec {\mathrm {i} }}(\mathbf {A} ^{\mathrm {S} })={\vec {0}} $

Bilinearform:

$ {\vec {u}}\cdot \mathbf {A} \cdot {\vec {v}}={\vec {v}}\cdot \mathbf {A} \cdot {\vec {u}}\quad \forall {\vec {u}},{\vec {v}}\in \mathbb {V} $

Alle #Eigenwerte λ1,2,3 sind reell. Alle #Eigenvektoren $ {\vec {a}}_{1,2,3} $ sind reell und paarweise orthogonal zueinander oder orthogonalisierbar. Hauptachsentransformation:

$ {\begin{aligned}\mathbf {A} =&\sum _{i=1}^{3}\lambda _{i}{\hat {a}}_{i}\otimes {\hat {a}}_{i}=({\hat {a}}_{i}\otimes {\hat {e}}_{i})\left(\sum _{i=1}^{3}\lambda _{j}{\hat {e}}_{j}\otimes {\hat {e}}_{j}\right)({\hat {e}}_{k}\otimes {\hat {a}}_{k})\\=&{\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}\cdot {\begin{pmatrix}\lambda _{1}&0&0\\0&\lambda _{2}&0\\0&0&\lambda _{3}\end{pmatrix}}\cdot {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}^{\top }\end{aligned}} $

Bezüglich der Standardbasis:

$ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}={\begin{pmatrix}A_{11}&A_{12}&A_{13}\\A_{12}&A_{22}&A_{23}\\A_{13}&A_{23}&A_{33}\end{pmatrix}} $

#Invarianten:

$ \mathrm {Sp} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{11}+A_{22}+A_{33} $
$ \mathrm {I} _{2}(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{11}A_{22}+A_{11}A_{33}+A_{22}A_{33}-A_{12}^{2}-A_{13}^{2}-A_{23}^{2} $
$ {\begin{aligned}\mathrm {det} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=&A_{11}(A_{22}A_{33}-A_{23}^{2})+A_{12}(A_{23}A_{13}-A_{12}A_{33})\\&+A_{13}(A_{12}A_{23}-A_{13}A_{22})\end{aligned}} $
$ \parallel A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\parallel ={\sqrt {A_{11}^{2}+A_{22}^{2}+A_{33}^{2}+2A_{12}^{2}+2A_{13}^{2}+2A_{23}^{2}}} $

Symmetrische und positiv definite Tensoren

Definition

$ \mathbf {A} :\quad \mathbf {A} =\mathbf {A} ^{\top }\quad {\text{und}}\quad {\vec {v}}\cdot \mathbf {A} \cdot {\vec {v}}>0\quad \forall \;{\vec {v}}\in \mathbb {V} \setminus \{{\vec {0}}\} $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathrm {adj} (\mathbf {A} )=\mathrm {det} (\mathbf {A} )\mathbf {A} ^{-1}=\mathbf {A} ^{2}-\mathrm {Sp} (\mathbf {A} )\mathbf {A} +\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} $

Mit den #Eigenwerten $ \lambda _{1},\,\lambda _{2},\,\lambda _{3} $, den #Eigenvektoren $ {\hat {a}}_{1},\,{\hat {a}}_{2},\,{\hat {a}}_{3} $ und einer reellwertigen Funktion $ f(x)\in \mathbb {R} $ eines reellen Argumentes $ x\in \mathbb {R} $ definiert man über das #Eigensystem symmetrischer Tensoren

$ {\begin{aligned}\mathbf {A} =&\sum _{i=1}^{3}\lambda _{i}{\hat {a}}_{i}\otimes {\hat {a}}_{i}\\=&{\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}\cdot {\begin{pmatrix}\lambda _{1}&0&0\\0&\lambda _{2}&0\\0&0&\lambda _{3}\end{pmatrix}}\cdot {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}^{\top }\end{aligned}} $

den Funktionswert des Tensors:

$ {\begin{aligned}f(\mathbf {A} ):=&\sum _{i=1}^{3}f(\lambda _{i}){\hat {a}}_{i}\otimes {\hat {a}}_{i}\\=&{\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}\cdot {\begin{pmatrix}f(\lambda _{1})&0&0\\0&f(\lambda _{2})&0\\0&0&f(\lambda _{3})\end{pmatrix}}\cdot {\begin{pmatrix}{\hat {a}}_{1}&{\hat {a}}_{2}&{\hat {a}}_{3}\end{pmatrix}}^{\top }\end{aligned}} $

Ist f eine mehrdeutige Funktion, wie die Wurzel (Mathematik), mit n alternativen Werten, dann steht f(A) mehrdeutig für n3 alternative Tensoren.

Insbesondere mit dem Deformationsgradient F:

Rechter Strecktensor

$ \mathbf {U} =+{\sqrt {\mathbf {F} ^{\top }\cdot \mathbf {F} }} $

Linker Strecktensor

$ \mathbf {v} =+{\sqrt {\mathbf {F\cdot F} ^{\top }}} $

Henky-Dehnung

$ \mathbf {E} _{H}:=\ln(\mathbf {U} )={\frac {1}{2}}\ln(\mathbf {F} ^{\top }\cdot \mathbf {F} ) $

Voigt-Notation symmetrischer Tensoren zweiter Stufe

Die Tensoren

$ {\begin{aligned}\mathbf {S} _{1}=&{\vec {e}}_{1}\otimes {\vec {e}}_{1}\\\mathbf {S} _{2}=&{\vec {e}}_{2}\otimes {\vec {e}}_{2}\\\mathbf {S} _{3}=&{\vec {e}}_{3}\otimes {\vec {e}}_{3}\\\mathbf {S} _{4}=&{\vec {e}}_{2}\otimes {\vec {e}}_{3}+{\vec {e}}_{3}\otimes {\vec {e}}_{2}\\\mathbf {S} _{5}=&{\vec {e}}_{1}\otimes {\vec {e}}_{3}+{\vec {e}}_{3}\otimes {\vec {e}}_{1}\\\mathbf {S} _{6}=&{\vec {e}}_{1}\otimes {\vec {e}}_{2}+{\vec {e}}_{2}\otimes {\vec {e}}_{1}\end{aligned}} $

bilden eine Basis im Vektorraum $ \mathrm {sym} (\mathbb {V} ,\mathbb {V} )\subset {\mathcal {L}} $ der symmetrischen Tensoren zweiter Stufe. Bezüglich dieser Basis können alle symmetrischen Tensoren zweiter Stufe in Voigt'scher Notation dargestellt werden:

$ \mathbf {A} \in \mathrm {sym} (\mathbb {V} ,\mathbb {V} )\quad \rightarrow \quad \mathbf {A} =A_{r}\mathbf {S} _{r}{\hat {=}}{\begin{bmatrix}A_{1}\\A_{2}\\A_{3}\\A_{4}\\A_{5}\\A_{6}\end{bmatrix}} $

Diese Vektoren dürfen addiert, subtrahiert und mit einem Skalar multipliziert werden. Beim Skalarprodukt muss

$ \mathbf {A} :\mathbf {B} =A_{1}B_{1}+A_{2}B_{2}+A_{3}B_{3}+2A_{4}B_{4}+2A_{5}B_{5}+2A_{6}B_{6} $

berücksichtigt werden. Siehe auch #Voigt'sche Notation von Tensoren vierter Stufe.

Schiefsymmetrische Tensoren

Definition

$ \mathbf {A} :\quad \mathbf {A} =-\mathbf {A} ^{\top } $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathrm {adj} (\mathbf {A} )=\mathbf {A\cdot A} +\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} =\mathbf {A\cdot A} -{\frac {1}{2}}\mathrm {Sp} (\mathbf {A} ^{2})\mathbf {1} $

#Invarianten:

$ \mathrm {Sp} (\mathbf {A} )=0 $
$ \mathrm {I} _{2}(\mathbf {A} )=-{\frac {1}{2}}\mathrm {Sp} (\mathbf {A} ^{2}) $
$ \mathrm {det} (\mathbf {A} )=0 $
$ \quad \parallel \mathbf {A} \parallel ={\sqrt {2\mathrm {I} _{2}(\mathbf {A} )}}={\sqrt {-\mathrm {Sp} (\mathbf {A} ^{2})}} $

In kartesischen Koordinaten:

$ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}={\begin{pmatrix}0&A_{12}&A_{13}\\-A_{12}&0&A_{23}\\-A_{13}&-A_{23}&0\end{pmatrix}} $

#Invarianten:

$ \mathrm {Sp} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=0 $
$ \mathrm {I} _{2}(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=A_{12}^{2}+A_{13}^{2}+A_{23}^{2} $
$ \mathrm {det} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=0 $
$ \parallel A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\parallel ={\sqrt {2}}{\sqrt {A_{12}^{2}+A_{13}^{2}+A_{23}^{2}}} $

Bilinearform:

$ {\vec {u}}\cdot \mathbf {A} \cdot {\vec {v}}=-{\vec {v}}\cdot \mathbf {A} \cdot {\vec {u}}\quad \forall {\vec {u}},{\vec {v}}\in \mathbb {V} $
$ {\vec {v}}\cdot \mathbf {A} \cdot {\vec {v}}=0\quad \forall {\vec {v}}\in \mathbb {V} $

Ein Eigenwert ist null, zwei imaginär konjugiert komplex, siehe #Axialer Tensor oder Kreuzproduktmatrix.

#Dualer axialer Vektor:

$ {\begin{aligned}&\mathbf {A} _{\times }:={\stackrel {A}{\overrightarrow {\mathbf {A} }}}:=-{\frac {1}{2}}\mathbf {1} \times \mathbf {A} ^{\top }=-{\frac {1}{2}}\mathbf {1} \cdot \!\!\times \mathbf {A} =-{\frac {1}{2}}{\vec {\mathrm {i} }}(\mathbf {A} )\\&\rightarrow \quad \mathbf {A} \cdot {\vec {v}}={\stackrel {A}{\overrightarrow {\mathbf {A} }}}\times {\vec {v}}\quad \forall {\vec {v}}\in \mathbb {V} \end{aligned}} $

mit #Vektorinvariante $ {\vec {\mathrm {i} }}(\mathbf {A} ) $. Der zum Eigenwert null gehörende #Eigenvektor ist proportional zum dualen axialen Vektor $ \mathbf {A} _{\times } $ denn

$ \mathbf {A\cdot A} _{\times }=\mathbf {A} _{\times }\times \mathbf {A} _{\times }={\vec {0}} $
$ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\quad \rightarrow \;\mathbf {A} _{\times }=-{\frac {1}{2}}A_{ij}{\hat {e}}_{i}\times {\hat {e}}_{j}={\begin{pmatrix}-A_{23}\\A_{13}\\-A_{12}\end{pmatrix}} $
$ \mathbf {A} =A_{ij}({\vec {a}}_{i}\otimes {\vec {b}}_{j}-{\vec {b}}_{j}\otimes {\vec {a}}_{i})\quad \rightarrow \;\mathbf {A} _{\times }=-A_{ij}{\vec {a}}_{i}\times {\vec {b}}_{j} $

Axialer Tensor oder Kreuzproduktmatrix

Kreuzproduktmatrix $ [{\vec {u}}]_{\times } $ eines Vektors $ {\vec {u}} $:

$ {\begin{aligned}{\vec {u}}=u_{i}{\hat {e}}_{i}=&{\begin{pmatrix}u_{1}\\u_{2}\\u_{3}\end{pmatrix}}\\\rightarrow \;[{\vec {u}}]_{\times }=&{\vec {u}}\times \mathbf {1} ={\vec {u}}\times {\hat {e}}_{i}\otimes {\hat {e}}_{i}=-{\stackrel {3}{\mathbf {E} }}\cdot {\vec {u}}={\begin{pmatrix}0&-u_{3}&u_{2}\\u_{3}&0&-u_{1}\\-u_{2}&u_{1}&0\end{pmatrix}}\in {\mathcal {L}}\end{aligned}} $

Kofaktor: $ \mathrm {cof} ({\vec {u}}\times \mathbf {1} )=\mathrm {adj} ({\vec {u}}\times \mathbf {1} )={\vec {u}}\otimes {\vec {u}} $

#Invarianten:

$ \mathrm {Sp} =0 $
$ \mathrm {I} _{2}={\vec {u}}\cdot {\vec {u}}=u_{1}^{2}+u_{2}^{2}+u_{3}^{2} $
$ \mathrm {det} =0 $
$ \|{\vec {u}}\times \mathbf {1} \|={\sqrt {2{\vec {u}}\cdot {\vec {u}}}}={\sqrt {2}}{\sqrt {u_{1}^{2}+u_{2}^{2}+u_{3}^{2}}} $
$ {\stackrel {A}{\overrightarrow {{\vec {u}}\times \mathbf {1} }}}={\vec {u}} $

#Eigensystem:

$ {\begin{aligned}\lambda _{1}=&0\,,&{\vec {v}}_{1}=&{\vec {u}}\\\lambda _{2,3}=&\mp \mathrm {i} |{\vec {u}}|\,,&{\vec {v}}_{2,3}&\simeq &{\frac {u_{1}}{|{\vec {u}}|}}{\begin{pmatrix}u_{1}\\u_{2}\\u_{3}\end{pmatrix}}\pm \mathrm {i} {\begin{pmatrix}\pm \mathrm {i} |{\vec {u}}|\\-u_{3}\\u_{2}\end{pmatrix}}\end{aligned}} $

Eigenschaften:

$ {\vec {u}}\times {\vec {v}}=({\vec {u}}\times \mathbf {1} )\cdot {\vec {v}}={\vec {u}}\cdot ({\vec {v}}\times \mathbf {1} ) $
$ {\vec {u}}\times \mathbf {1} =\mathbf {1} \times {\vec {u}} $
$ ({\vec {u}}\times \mathbf {1} )^{\top }=-{\vec {u}}\times \mathbf {1} $
$ {\vec {u}}=-{\frac {1}{2}}\mathbf {1} \cdot \!\!\times ({\vec {u}}\times \mathbf {1} )=-{\frac {1}{2}}(\mathbf {1} \times {\vec {u}})\times \mathbf {1} ={\frac {1}{2}}\mathbf {1} \times ({\vec {u}}\times \mathbf {1} ) $
$ {\vec {u}}\times ({\vec {v}}\times \mathbf {1} )=({\vec {u}}\times \mathbf {1} )\cdot ({\vec {v}}\times \mathbf {1} )={\vec {v}}\otimes {\vec {u}}-({\vec {u}}\cdot {\vec {v}})\mathbf {1} $
$ {\vec {u}}\times ({\vec {v}}\times \mathbf {1} )\cdot {\vec {w}}={\vec {u}}\times ({\vec {v}}\times {\vec {w}})=({\vec {u}}\cdot {\vec {w}}){\vec {v}}-({\vec {u}}\cdot {\vec {v}}){\vec {w}} $

Potenzen von $ [{\vec {u}}]_{\times }={\vec {u}}\times \mathbf {1} $

$ [{\vec {u}}]_{\times }^{2}=[{\vec {u}}]_{\times }\cdot [{\vec {u}}]_{\times }={\vec {u}}\otimes {\vec {u}}-({\vec {u}}\cdot {\vec {u}})\mathbf {1} $
$ [{\vec {u}}]_{\times }^{3}=-({\vec {u}}\cdot {\vec {u}})[{\vec {u}}]_{\times } $

Deviatorische Tensoren

Definition

$ \mathbf {A} :\quad \mathrm {Sp} (\mathbf {A} )=0 $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\left(\mathbf {A} ^{2}\right)^{\top }+\mathrm {I} _{2}(\mathbf {A} )\mathbf {1} =\left(\mathbf {A} ^{2}\right)^{\top }-{\frac {1}{2}}\mathrm {Sp} (\mathbf {A} ^{2})\mathbf {1} $

#Hauptinvarianten:

$ \mathrm {Sp} (\mathbf {A} ):=0 $
$ \mathrm {I} _{2}(\mathbf {A} )=-{\frac {1}{2}}\mathrm {Sp} (\mathbf {A} ^{2}) $
$ \mathrm {det} (\mathbf {A} )={\frac {1}{3}}\mathrm {Sp} (\mathbf {A} ^{3}) $

Bezüglich der Standardbasis:

$ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}={\begin{pmatrix}A_{11}&A_{12}&A_{13}\\A_{21}&A_{22}&A_{23}\\A_{31}&A_{32}&-A_{11}-A_{22}\end{pmatrix}} $
$ \mathrm {Sp} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=0 $
$ \mathrm {I} _{2}(A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=-A_{11}^{2}-A_{22}^{2}-A_{11}A_{22}-A_{12}A_{21}-A_{13}A_{31}-A_{23}A_{32} $
$ {\begin{aligned}\mathrm {det} (A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j})=&-A_{11}(A_{11}A_{22}+A_{22}^{2}+A_{23}A_{32})\\&+A_{12}(A_{23}A_{31}+A_{21}A_{11}+A_{21}A_{22})\\&+A_{13}(A_{21}A_{32}-A_{22}A_{31})\end{aligned}} $
$ \parallel A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j}\parallel ={\sqrt {\begin{array}{r}2A_{11}^{2}+2A_{22}^{2}+2A_{11}A_{22}+A_{12}^{2}+A_{21}^{2}+\ldots \\\ldots +A_{13}^{2}+A_{31}^{2}+A_{23}^{2}+A_{32}^{2}\end{array}}} $

Kugeltensoren

Definition

$ \mathbf {A} :\quad \mathbf {A} =a\mathbf {1} ={\begin{pmatrix}a&0&0\\0&a&0\\0&0&a\end{pmatrix}} $

Kofaktor: $ \mathrm {cof} (\mathbf {A} )=\mathrm {adj} (\mathbf {A} )=a^{2}\mathbf {1} $

$ \mathrm {Sp} (\mathbf {A} )=3a $
$ \mathrm {I} _{2}(\mathbf {A} )=3a^{2} $
$ \mathrm {det} (\mathbf {A} )=a^{3} $
$ \parallel \mathbf {A} \parallel ={\sqrt {3}}|a| $

Dekompositionen eines Tensors

Gegeben ein beliebiger Tensor $ \mathbf {A} =A_{ij}{\hat {e}}_{i}\otimes {\hat {e}}_{j} $

Symmetrischer Anteil

$ \mathbf {A} ^{\mathrm {S} }=\mathrm {sym} (\mathbf {A} ):={\frac {1}{2}}(\mathbf {A} +\mathbf {A} ^{\top }) $
$ \mathbf {A} ^{\mathrm {S} }={\frac {1}{2}}{\begin{pmatrix}2A_{11}&A_{12}+A_{21}&A_{13}+A_{31}\\A_{12}+A_{21}&2A_{22}&A_{23}+A_{32}\\A_{13}+A_{31}&A_{23}+A_{32}&2A_{33}\end{pmatrix}} $
$ \mathrm {Sp} (\mathbf {A} ^{\mathrm {S} })=\mathrm {Sp} (\mathbf {A} )=A_{11}+A_{22}+A_{33} $
$ {\begin{aligned}\mathrm {I} _{2}(\mathbf {A} ^{\mathrm {S} })=&{\frac {1}{2}}\mathrm {I} _{2}(\mathbf {A} )+{\frac {1}{4}}\mathrm {Sp} ^{2}(\mathbf {A} )-{\frac {1}{4}}\mathbf {A:A} \\=&A_{11}A_{22}+A_{11}A_{33}+A_{22}A_{33}\\&-{\frac {1}{4}}\left[(A_{12}+A_{21})^{2}+(A_{13}+A_{31})^{2}+(A_{23}+A_{32})^{2}\right]\end{aligned}} $
$ {\begin{aligned}\mathrm {det} (\mathbf {A} ^{\mathrm {S} })=&{\frac {1}{4}}\mathrm {det} (\mathbf {A} )+{\frac {1}{4}}\mathbf {A} :\mathrm {adj} (\mathbf {A} )\\=&A_{11}A_{22}A_{33}+{\frac {1}{4}}(A_{12}+A_{21})(A_{23}+A_{32})(A_{13}+A_{31})\\&-{\frac {1}{4}}\left[A_{11}(A_{23}+A_{32})^{2}+A_{22}(A_{13}+A_{31})^{2}+A_{33}(A_{12}+A_{21})^{2}\right]\end{aligned}} $
$ {\begin{aligned}\parallel (\mathbf {A} ^{\mathrm {S} })\parallel =&{\sqrt {\mathbf {A:A} ^{\mathrm {S} }}}\\=&{\sqrt {A_{11}^{2}+A_{22}^{2}+A_{33}^{2}+{\frac {1}{2}}[(A_{12}+A_{21})^{2}+(A_{13}+A_{31})^{2}+(A_{23}+A_{32})^{2}]}}\end{aligned}} $

Schiefsymmetrischer Anteil

$ \mathbf {A} ^{\mathrm {A} }=\mathrm {skw} (\mathbf {A} ):={\frac {1}{2}}(\mathbf {A} -\mathbf {A} ^{\top }) $
$ \mathbf {A} ^{\mathrm {A} }={\frac {1}{2}}{\begin{pmatrix}0&A_{12}-A_{21}&A_{13}-A_{31}\\A_{21}-A_{12}&0&A_{23}-A_{32}\\A_{31}-A_{13}&A_{32}-A_{23}&0\end{pmatrix}} $
$ \mathrm {Sp} (\mathbf {A} ^{\mathrm {A} })=0 $
$ {\begin{aligned}\mathrm {I} _{2}(\mathbf {A} ^{\mathrm {A} })=&{\frac {1}{4}}\left[\mathbf {A:A} -\mathrm {Sp} (\mathbf {A} ^{2})\right]\\=&{\frac {1}{4}}\left[(A_{12}-A_{21})^{2}+(A_{13}-A_{31})^{2}+(A_{23}-A_{32})^{2}\right]\end{aligned}} $
$ \mathrm {det} (\mathbf {A} ^{\mathrm {A} })=0 $
$ \parallel \mathbf {A} ^{\mathrm {A} }\parallel ={\sqrt {\mathbf {A:A} ^{\mathrm {A} }}}={\sqrt {\frac {1}{2}}}{\sqrt {(A_{12}-A_{21})^{2}+(A_{13}-A_{31})^{2}+(A_{32}-A_{23})^{2}}} $

Deviator

$ \mathbf {A} ^{\mathrm {D} }=\mathrm {dev} (\mathbf {A} ):=\mathbf {A} -{\frac {1}{3}}\mathrm {Sp} (\mathbf {A} )\mathbf {1} $
$ \mathbf {A} ^{\mathrm {D} }={\begin{pmatrix}{\frac {2A_{11}-A_{22}-A_{33}}{3}}&A_{12}&A_{13}\\A_{21}&{\frac {2A_{22}-A_{11}-A_{33}}{3}}&A_{23}\\A_{31}&A_{32}&{\frac {2A_{33}-A_{11}-A_{22}}{3}}\end{pmatrix}} $
$ \mathrm {Sp} (\mathbf {A} ^{\mathrm {D} })=0 $
$ {\begin{aligned}\mathrm {I} _{2}(\mathbf {A} ^{\mathrm {D} })=&\mathrm {I} _{2}(\mathbf {A} )-{\frac {1}{3}}\mathrm {Sp} ^{2}(\mathbf {A} )={\frac {1}{6}}\mathrm {Sp} ^{2}(\mathbf {A} )-{\frac {1}{2}}\mathrm {Sp} (\mathbf {A} ^{2})\\=&{\frac {1}{3}}(A_{11}A_{22}+A_{11}A_{33}+A_{22}A_{33}-A_{11}^{2}-A_{22}^{2}-A_{33}^{2})\\&-A_{12}A_{21}-A_{13}A_{31}-A_{23}A_{32}\end{aligned}} $
$ {\begin{aligned}\mathrm {det} (\mathbf {A} ^{\mathrm {D} })=&\mathrm {det} (\mathbf {A} )+{\frac {2}{27}}\mathrm {Sp} ^{3}(\mathbf {A} )-{\frac {1}{3}}\mathrm {Sp} (\mathbf {A} )\mathrm {I} _{2}(\mathbf {A} )\\=&{\frac {1}{27}}{\Big [}12A_{11}A_{22}A_{33}+2(A_{11}^{3}+A_{22}^{3}+A_{33}^{3})\ldots \\&\qquad \ldots -3A_{11}^{2}(A_{22}+A_{33})-3A_{22}^{2}(A_{11}+A_{33})-3A_{33}^{2}(A_{11}+A_{22}){\Big ]}\\&-{\frac {1}{3}}{\Big [}(2A_{11}-A_{22}-A_{33})A_{23}A_{32}+(2A_{22}-A_{11}-A_{33})A_{13}A_{31}+\ldots \\&\qquad \ldots +(2A_{33}-A_{11}-A_{22})A_{12}A_{21}{\Big ]}\\&+A_{13}A_{32}A_{21}+A_{12}A_{23}A_{31}\end{aligned}} $
$ \parallel \mathbf {A} ^{\mathrm {D} }\parallel ={\sqrt {\begin{array}{r}{\frac {2}{3}}(A_{11}^{2}+A_{22}^{2}+A_{33}^{2}-A_{11}A_{22}-A_{11}A_{33}-A_{22}A_{33})+\ldots \\\ldots +A_{12}^{2}+A_{21}^{2}+A_{13}^{2}+A_{31}^{2}+A_{23}^{2}+A_{32}^{2}\end{array}}} $

Kugelanteil

$ \mathbf {A} ^{\mathrm {K} }=\mathrm {sph} (\mathbf {A} ):={\frac {1}{3}}\mathrm {Sp} (\mathbf {A} )\mathbf {1} $
$ \mathbf {A} ^{\mathrm {K} }={\frac {1}{3}}(A_{11}+A_{22}+A_{33}){\begin{pmatrix}1&0&0\\0&1&0\\0&0&1\end{pmatrix}} $
$ \mathrm {Sp} (\mathbf {A} ^{\mathrm {K} })=\mathrm {Sp} (\mathbf {A} )=A_{11}+A_{22}+A_{33} $
$ \mathrm {I} _{2}(\mathbf {A} ^{\mathrm {K} })={\frac {1}{3}}\mathrm {Sp} ^{2}(\mathbf {A} )={\frac {1}{3}}(A_{11}+A_{22}+A_{33})^{2} $
$ \mathrm {det} (\mathbf {A} ^{\mathrm {K} })={\frac {1}{27}}\mathrm {Sp} ^{3}(\mathbf {A} )={\frac {1}{27}}(A_{11}+A_{22}+A_{33})^{3} $
$ \parallel \mathbf {A} ^{\mathrm {K} }\parallel ={\frac {1}{\sqrt {3}}}|\mathrm {Sp} (\mathbf {A} )|={\frac {1}{\sqrt {3}}}|A_{11}+A_{22}+A_{33}| $

Beziehung zwischen den Anteilen des Tensors

$ \mathbf {A} =\mathbf {A} ^{\mathrm {S} }+\mathbf {A} ^{\mathrm {A} }=\mathbf {A} ^{\mathrm {D} }+\mathbf {A} ^{\mathrm {K} } $

Symmetrische und schiefsymmetrische Tensoren sind orthogonal zueinander:

$ \mathbf {A} ^{\mathrm {S} }:\mathbf {B} ^{\mathrm {A} }=0 $

Deviatoren und Kugeltensoren sind orthogonal zueinander:

$ \mathbf {A} ^{\mathrm {D} }:\mathbf {B} ^{\mathrm {K} }=0 $

Polarzerlegung

Für jeden Tensor F mit #Determinante ≠ 0 gibt es #Orthogonale Tensoren Q und #Symmetrische und positiv definite Tensoren U in eindeutiger Weise, sodass

F = Q·U

Im Fall des Deformationsgradienten ist U der rechte Strecktensor, siehe #Symmetrische und positiv definite Tensoren. Der Anteil U berechnet sich wie dort angegeben aus

$ \mathbf {U} =+{\sqrt {\mathbf {F^{\top }\cdot F} }} $

Dann ist U·U = F·F und

$ \mathbf {Q} =\mathbf {F\cdot U} ^{-1} $

Bei det(F)=0 ergeben sich U sowie Q aus der Singulärwertzerlegung von F und U ist nur noch symmetrisch positiv semidefinit.

Projektionen

Punkt auf Gerade

Gegeben sei die Gerade durch den Punkt $ {\vec {x}} $ mit Richtungsvektor $ {\vec {g}} $ und ein beliebiger anderer Punkt $ {\vec {p}} $.

Dann ist

$ {\begin{aligned}{\vec {p}}=&{\vec {x}}+{\vec {a}}+{\vec {b}}\quad {\textsf {mit}}\quad {\vec {a}}\|{\vec {g}}\quad {\text{und}}\quad {\vec {b}}\bot {\vec {g}}\\\mathbf {G} =&{\frac {{\vec {g}}\otimes {\vec {g}}}{{\vec {g}}\cdot {\vec {g}}}}\quad \rightarrow \quad \mathbf {G} \cdot {\vec {g}}={\vec {g}}\,,\quad (\mathbf {1} -\mathbf {G} )\cdot {\vec {g}}={\vec {0}}\\&{\vec {n}}\cdot {\vec {g}}=0\quad \rightarrow \quad \mathbf {G} \cdot {\vec {n}}={\vec {0}}\,,\quad (\mathbf {1} -\mathbf {G} )\cdot {\vec {n}}={\vec {n}}\\{\vec {a}}=&\mathbf {G} \cdot ({\vec {p}}-{\vec {x}})={\frac {{\vec {g}}\cdot ({\vec {p}}-{\vec {x}})}{{\vec {g}}\cdot {\vec {g}}}}{\vec {g}}\\{\vec {b}}=&\left(\mathbf {1} -\mathbf {G} \right)\cdot ({\vec {p}}-{\vec {x}})={\vec {p}}-{\vec {x}}-{\vec {a}}\end{aligned}} $

Der Punkt $ {\vec {x}}+{\vec {a}} $ ist die senkrechte Projektion von $ {\vec {p}} $ auf die Gerade. Der Tensor G extrahiert den Anteil eines Vektors in Richtung von $ {\vec {g}} $ und 1-G den Anteil senkrecht dazu.

Punkt oder Gerade auf Ebene

Gegeben sei die Ebene durch den Punkt $ {\vec {x}} $ und zwei die Ebene aufspannende Vektoren $ {\vec {u}} $ und $ {\vec {v}}\not \!\|{\vec {u}} $ sowie ein beliebiger anderer Punkt $ {\vec {p}} $. Dann verschwindet die Normale

$ {\hat {n}}={\frac {{\vec {u}}\times {\vec {v}}}{|{\vec {u}}\times {\vec {v}}|}} $

nicht. Dann ist

$ {\begin{aligned}{\vec {p}}=&{\vec {x}}+{\vec {a}}+{\vec {b}}\quad {\textsf {mit}}\quad {\vec {a}}\bot {\hat {n}}\quad {\text{und}}\quad {\vec {b}}\|{\hat {n}}\\\mathbf {P} =&{\frac {({\vec {v}}\cdot {\vec {v}}){\vec {u}}\otimes {\vec {u}}-({\vec {u}}\cdot {\vec {v}})({\vec {u}}\otimes {\vec {v}}+{\vec {v}}\otimes {\vec {u}})+({\vec {u}}\cdot {\vec {u}}){\vec {v}}\otimes {\vec {v}}}{({\vec {u}}\cdot {\vec {u}})({\vec {v}}\cdot {\vec {v}})-({\vec {u}}\cdot {\vec {v}})^{2}}}=\mathbf {1} -{\hat {n}}\otimes {\hat {n}}\\&\rightarrow \mathbf {P} \cdot {\vec {u}}={\vec {u}}\,,\quad \mathbf {P} \cdot {\vec {v}}={\vec {v}}\,,\quad \mathbf {P} \cdot {\hat {n}}={\vec {0}}\,,\quad (\mathbf {1} -\mathbf {P} )\cdot {\hat {n}}={\hat {n}}\\&\rightarrow \mathbf {P} \cdot (x{\vec {u}}+y{\vec {v}})=x{\vec {u}}+y{\vec {v}}\quad {\text{und}}\quad (\mathbf {1} -\mathbf {P} )\cdot (x{\vec {u}}+y{\vec {v}})={\vec {0}}\quad \forall x,y\in \mathbb {R} \\{\vec {a}}=&\mathbf {P} \cdot ({\vec {p}}-{\vec {x}})\\{\vec {b}}=&(\mathbf {1} -\mathbf {P} )\cdot ({\vec {p}}-{\vec {x}})={\vec {p}}-{\vec {x}}-{\vec {a}}\end{aligned}} $

Der Punkt $ {\vec {x}}+{\vec {a}} $ ist die senkrechte Projektion von $ {\vec {p}} $ auf die Ebene.[2] Der Tensor P extrahiert den Anteil eines Vektors in der Ebene und 1-P den Anteil senkrecht dazu.

Die Projektion der Geraden, die durch die Punkte $ {\vec {x}} $ und $ {\vec {p}} $ verläuft, liegt in der Ebene in Richtung des Vektors $ {\vec {a}} $.

Falls $ |{\vec {u}}|=|{\vec {v}}|=1 $ und $ {\vec {u}}\bot {\vec {v}} $ folgt:

$ {\hat {n}}={\vec {u}}\times {\vec {v}}\quad {\text{mit}}\quad |{\hat {n}}|=1 $
$ \mathbf {P} ={\vec {u}}\otimes {\vec {u}}+{\vec {v}}\otimes {\vec {v}}=\mathbf {1} -{\hat {n}}\otimes {\hat {n}} $
$ {\vec {a}}=({\vec {u}}\otimes {\vec {u}}+{\vec {v}}\otimes {\vec {v}})\cdot ({\vec {p}}-{\vec {x}})=(\mathbf {1} -{\hat {n}}\otimes {\hat {n}})\cdot ({\vec {p}}-{\vec {x}}) $
$ {\vec {b}}=(\mathbf {1} -{\vec {u}}\otimes {\vec {u}}-{\vec {v}}\otimes {\vec {v}})\cdot ({\vec {p}}-{\vec {x}})=({\hat {n}}\otimes {\hat {n}})\cdot ({\vec {p}}-{\vec {x}}) $

Fundamentaltensor 3. Stufe

Definition:

$ {\begin{aligned}{\stackrel {3}{\mathbf {E} }}:=&\epsilon _{ijk}\,{\hat {e}}_{i}\otimes {\hat {e}}_{j}\otimes {\hat {e}}_{k}\\=&({\hat {e}}_{j}\times {\hat {e}}_{k})\otimes {\hat {e}}_{j}\otimes {\hat {e}}_{k}\\=&{\hat {e}}_{i}\otimes ({\hat {e}}_{k}\times {\hat {e}}_{i})\otimes {\hat {e}}_{k}\\=&{\hat {e}}_{i}\otimes {\hat {e}}_{j}\otimes ({\hat {e}}_{i}\times {\hat {e}}_{j})\end{aligned}} $

Kreuzprodukt von Vektoren:

$ {\vec {u}}\times {\vec {v}}={\stackrel {3}{\mathbf {E} }}:({\vec {u}}\otimes {\vec {v}})={\vec {v}}\cdot {\stackrel {3}{\mathbf {E} }}\cdot {\vec {u}}=-{\vec {u}}\cdot {\stackrel {3}{\mathbf {E} }}\cdot {\vec {v}}=-{\stackrel {3}{\mathbf {E} }}:({\vec {v}}\otimes {\vec {u}})=-{\vec {v}}\times {\vec {u}} $
$ {\vec {e}}_{i}\times {\vec {e}}_{j}=\epsilon _{ijk}\,{\hat {e}}_{k} $

#Kreuzprodukt von Tensoren, #Skalarkreuzprodukt von Tensoren:

$ {\begin{aligned}{\stackrel {3}{\mathbf {E} }}:\mathbf {A} =&\mathbf {A} :{\stackrel {3}{\mathbf {E} }}=-{\stackrel {3}{\mathbf {E} }}:(\mathbf {A} ^{\top })=-(\mathbf {A} ^{\top }):{\stackrel {3}{\mathbf {E} }}\\=&\mathbf {1} \times \mathbf {A} ^{\top }=\mathbf {1} \cdot \!\!\times \mathbf {A} \end{aligned}} $

#Dualer axialer Vektor und #Vektorinvariante:

$ {\stackrel {3}{\mathbf {E} }}:\mathbf {A} =-2{\stackrel {A}{\overrightarrow {\mathbf {A} }}}={\vec {\mathrm {i} }}(\mathbf {A} ) $

#Kreuzprodukt von Tensoren:

$ \mathbf {A} \times \mathbf {B} ={\stackrel {3}{\mathbf {E} }}:(\mathbf {A\cdot B} ^{\top }) $
$ (A_{ik}{\vec {e}}_{i}\otimes {\vec {e}}_{k})\times (B_{jl}{\vec {e}}_{j}\otimes {\vec {e}}_{l})=A_{ik}B_{jk}{\vec {e}}_{i}\times {\vec {e}}_{j}=\epsilon _{ijk}A_{jl}B_{kl}{\vec {e}}_{i} $

#Skalarkreuzprodukt von Tensoren:

$ \mathbf {A} \cdot \!\!\times \mathbf {B} ={\stackrel {3}{\mathbf {E} }}:(\mathbf {A\cdot B} ) $
$ (A_{ik}{\vec {e}}_{i}\otimes {\vec {e}}_{k})\cdot \!\!\times (B_{lj}{\vec {e}}_{l}\otimes {\vec {e}}_{j})=A_{ik}B_{kj}{\vec {e}}_{i}\times {\vec {e}}_{j}=\epsilon _{ijk}A_{jl}B_{lk}{\vec {e}}_{i} $

#Axialer Tensor oder Kreuzproduktmatrix:

$ {\stackrel {3}{\mathbf {E} }}\cdot {\vec {u}}={\vec {u}}\cdot {\stackrel {3}{\mathbf {E} }}=-{\vec {u}}\times \mathbf {1} =-\mathbf {1} \times {\vec {u}} $

Tensoren vierter Stufe

Tensoren zweiter Stufe sind ebenfalls Elemente eines Vektorraums $ {\mathcal {L}} $ wie im Abschnitt #Tensoren als Elemente eines Vektorraumes dargestellt. Daher kann man Tensoren vierter Stufe definieren, indem man in dem Kapitel formal die Tensoren zweiter Stufe durch Tensoren vierter Stufe und die Vektoren durch Tensoren zweiter Stufe ersetzt, z. B.:

$ {\stackrel {4}{\mathbf {A} }}=A_{pq}(\mathbf {A} _{p}\otimes \mathbf {G} _{q}) $

mit Komponenten $ A_{pq} $ und die Tensoren $ \mathbf {A} _{1},\mathbf {A} _{2},\ldots ,\mathbf {A} _{9}\in {\mathcal {L}} $ sowie $ \mathbf {G} _{1},\mathbf {G} _{2},\ldots ,\mathbf {G} _{9}\in {\mathcal {L}} $ bilden eine Basis von $ {\mathcal {L}} $.

Standardbasis in $ {\mathcal {L}} $:

$ \mathbf {E} _{1}={\vec {e}}_{1}\otimes {\vec {e}}_{1},\mathbf {E} _{2}={\vec {e}}_{1}\otimes {\vec {e}}_{2},\mathbf {E} _{3}={\vec {e}}_{1}\otimes {\vec {e}}_{3},\mathbf {E} _{4}={\vec {e}}_{2}\otimes {\vec {e}}_{1},\ldots ,\mathbf {E} _{9}={\vec {e}}_{3}\otimes {\vec {e}}_{3} $

Tensortransformation:

$ {\stackrel {4}{\mathbf {A} }}:\mathbf {H} =A_{pq}(\mathbf {A} _{p}\otimes \mathbf {G} _{q}):\mathbf {H} :=A_{pq}(\mathbf {G} _{q}:\mathbf {H} )\mathbf {A} _{p} $

Tensorprodukt:

$ [A_{pq}(\mathbf {A} _{p}\otimes \mathbf {G} _{q})]:[B_{rs}(\mathbf {H} _{r}\otimes \mathbf {U} _{s})]:=A_{pq}(\mathbf {G} _{q}:\mathbf {H} _{r})B_{rs}\mathbf {A} _{p}\otimes \mathbf {U} _{s} $

Übliche Schreibweisen für Tensoren vierter Stufe:

$ {\stackrel {4}{\mathbf {A} }}=\mathbb {A} =A_{ijkl}\;{\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l} $

Transpositionen

Transposition:

$ (\mathbf {A} \otimes \mathbf {B} )^{\top }=\mathbf {B} \otimes \mathbf {A} $
$ (A_{ijkl}\;{\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})^{\top }:=A_{ijkl}\;{\vec {e}}_{k}\otimes {\vec {e}}_{l}\otimes {\vec {e}}_{i}\otimes {\vec {e}}_{j} $

Spezielle Transposition $ {\stackrel {4}{\mathbf {A} }}{}^{\stackrel {mn}{\top }} $ vertauscht $ m $-tes mit $ n $-tem Basissystem.

Beispielsweise:

$ {\stackrel {4}{\mathbf {A} }}{}^{\stackrel {13}{\top }}:=A_{ijkl}\;{\vec {e}}_{k}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{i}\otimes {\vec {e}}_{l} $
$ {\stackrel {4}{\mathbf {A} }}{}^{\stackrel {24}{\top }}:=A_{ijkl}\;{\vec {e}}_{i}\otimes {\vec {e}}_{l}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{j} $
$ {\stackrel {4}{\mathbf {A} }}\,^{\top }=\left({\stackrel {4}{\mathbf {A} }}{}^{\stackrel {13}{\top }}\right){}^{\stackrel {24}{\top }}=A_{ijkl}\;{\vec {e}}_{k}\otimes {\vec {e}}_{l}\otimes {\vec {e}}_{i}\otimes {\vec {e}}_{j} $

Symmetrische Tensoren vierter Stufe

Definition: $ {\stackrel {4}{\mathbf {A} }}={\stackrel {4}{\mathbf {A} }}{}^{\top } $

Dann gilt: $ {\stackrel {4}{\mathbf {A} }}:\mathbf {B} =\mathbf {B} :{\stackrel {4}{\mathbf {A} }} $

Einheitstensor vierter Stufe

$ {\begin{aligned}{\stackrel {4}{\mathbf {1} }}:=&\mathbf {E} _{p}\otimes \mathbf {E} _{p}={\stackrel {4}{\mathbf {1} }}{}^{\top }=(\mathbf {1} \otimes \mathbf {1} )\,^{\stackrel {23}{\top }}\\=&{\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{i}\otimes {\vec {e}}_{j}=\delta _{ik}\delta _{jl}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\end{aligned}} $

Spezielle Tensoren vierter Stufe

Für beliebige Tensoren zweiter Stufe A gilt:

$ {\stackrel {4}{\mathbf {C} }}=\mathbf {E} _{p}^{\top }\otimes \mathbf {E} _{p}=\delta _{il}\delta _{jk}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {A} =\mathbf {A} ^{\top } $
$ {\stackrel {4}{\mathbf {C} }}={\frac {1}{3}}\mathbf {1} \otimes \mathbf {1} ={\frac {1}{3}}\delta _{ij}\delta _{kl}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {A} =\mathbf {A} ^{\mathrm {K} } $
$ {\stackrel {4}{\mathbf {C} }}={\stackrel {4}{\mathbf {1} }}-{\frac {1}{3}}\mathbf {1} \otimes \mathbf {1} =(\delta _{ik}\delta _{jl}-{\frac {1}{3}}\delta _{ij}\delta _{kl})({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {A} =\mathbf {A} ^{\mathrm {D} } $
$ {\stackrel {4}{\mathbf {C} }}={\frac {1}{2}}\left({\stackrel {4}{\mathbf {1} }}+\mathbf {E} _{p}^{\top }\otimes \mathbf {E} _{p}\right)={\frac {1}{2}}(\delta _{ik}\delta _{jl}+\delta _{il}\delta _{jk})({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {A} =\mathbf {A} ^{\mathrm {S} } $
$ {\stackrel {4}{\mathbf {C} }}={\frac {1}{2}}\left({\stackrel {4}{\mathbf {1} }}-\mathbf {E} _{p}^{\top }\otimes \mathbf {E} _{p}\right)={\frac {1}{2}}(\delta _{ik}\delta _{jl}-\delta _{il}\delta _{jk})({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {A} =\mathbf {A} ^{\mathrm {A} } $

Diese fünf Tensoren sind sämtlich symmetrisch.

Mit beliebigen Tensoren zweiter Stufe A, B und G gilt:

$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} \otimes \mathbf {B} ^{\top })^{\stackrel {23}{\top }}=A_{ik}B_{lj}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A\cdot G\cdot B} $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} ^{\top }\otimes \mathbf {B} ^{\top })^{\stackrel {23}{\top }}=A_{ki}B_{lj}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A} ^{\top }\cdot \mathbf {G\cdot B} $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} \otimes \mathbf {B} )^{\stackrel {23}{\top }}=A_{ik}B_{jl}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A\cdot G\cdot B} ^{\top } $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} ^{\top }\otimes \mathbf {B} )^{\stackrel {23}{\top }}=A_{ki}B_{jl}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A} ^{\top }\cdot \mathbf {G\cdot B} ^{\top } $

In dem in diesen Formeln im Tensor vierter Stufe B durch B und die Transpositionen $ {\stackrel {23}{\top }} $ durch $ {\stackrel {24}{\top }} $ ersetzt werden, entstehen die Ergebnisse mit transponiertem G:

$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} \otimes \mathbf {B} )^{\stackrel {24}{\top }}=A_{il}B_{kj}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A\cdot G} ^{\top }\cdot \mathbf {B} $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} ^{\top }\otimes \mathbf {B} )^{\stackrel {24}{\top }}=A_{li}B_{kj}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A} ^{\top }\cdot \mathbf {G} ^{\top }\cdot \mathbf {B} $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} \otimes \mathbf {B} ^{\top })^{\stackrel {24}{\top }}=A_{il}B_{jk}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A\cdot G} ^{\top }\cdot \mathbf {B} ^{\top } $
$ {\stackrel {4}{\mathbf {C} }}=(\mathbf {A} ^{\top }\otimes \mathbf {B} ^{\top })^{\stackrel {24}{\top }}=A_{li}B_{jk}({\vec {e}}_{i}\otimes {\vec {e}}_{j}\otimes {\vec {e}}_{k}\otimes {\vec {e}}_{l})\rightarrow {\stackrel {4}{\mathbf {C} }}:\mathbf {G} =\mathbf {A} ^{\top }\cdot \mathbf {G} ^{\top }\cdot \mathbf {B} ^{\top } $

Invertierungsformel

$ \left(a{\stackrel {4}{\mathbf {1} }}+\mathbf {B} \otimes \mathbf {C} \right)^{-1}={\frac {1}{a}}\left({\stackrel {4}{\mathbf {1} }}-{\frac {1}{a+\mathbf {B} :\mathbf {C} }}\mathbf {B} \otimes \mathbf {C} \right) $

Hooke'sches Gesetz

Mit den Spannungen $ \mathbf {T} $ und den Dehnungen $ \mathbf {E} $ im Hooke'schen Gesetz gilt:

$ {\stackrel {4}{\mathbf {C} }}:=2\mu {\stackrel {4}{\mathbf {1} }}+\lambda \mathbf {1} \otimes \mathbf {1} \quad \rightarrow \quad {\stackrel {4}{\mathbf {C} }}:\mathbf {E} =\mathbf {T} $

mit den Lamé-Konstanten $ \lambda $ und $ \mu $. Dieser Elastizitätstensor ist symmetrisch.

Invertierungsformel mit $ a=2\mu $, $ \mathbf {B} =\lambda \mathbf {1} $ und $ \mathbf {C} =\mathbf {1} $:

$ {\begin{aligned}&{\stackrel {4}{\mathbf {S} }}:={\stackrel {4}{\mathbf {C} }}{}^{-1}={\frac {1}{2\mu }}\left({\stackrel {4}{\mathbf {1} }}-{\frac {\lambda }{2\mu +3\lambda }}\mathbf {1} \otimes \mathbf {1} \right)={\frac {1}{2\mu }}{\stackrel {4}{\mathbf {1} }}-{\frac {\nu }{E}}\mathbf {1} \otimes \mathbf {1} \\&\rightarrow \quad {\stackrel {4}{\mathbf {S} }}:\mathbf {T} =\mathbf {E} \end{aligned}} $

mit der Querdehnzahl $ \nu $ und dem Elastizitätsmodul $ E $.

Voigt'sche Notation von Tensoren vierter Stufe

Aus der Basis $ \mathbf {S} _{1},\ldots ,\mathbf {S} _{6} $ des Vektorraums $ {\mathcal {S}}=\mathrm {sym} (\mathbb {V} ,\mathbb {V} ) $ der symmetrischen Tensoren zweiter Stufe, siehe #Voigt-Notation symmetrischer Tensoren zweiter Stufe, kann eine Basis des Vektorraums $ {\stackrel {4}{\mathcal {S}}}=\mathrm {Lin} ({\mathcal {S}},{\mathcal {S}}) $ der linearen Abbildungen von symmetrischen Tensoren auf symmetrische Tensoren konstruiert werden. Die 36 Komponenten der Tensoren vierter Stufe aus $ {\stackrel {4}{\mathcal {S}}} $ können als Voigt'scher Notation in eine 6×6-Matrix einsortiert werden:

$ {\stackrel {4}{\mathbf {A} }}=A_{uv}\mathbf {S} _{u}\otimes \mathbf {S} _{v}{\hat {=}}{\begin{bmatrix}A_{11}&A_{12}&A_{13}&A_{14}&A_{15}&A_{16}\\A_{21}&A_{22}&A_{23}&A_{24}&A_{25}&A_{26}\\A_{31}&A_{32}&A_{33}&A_{34}&A_{35}&A_{36}\\A_{41}&A_{42}&A_{43}&A_{44}&A_{45}&A_{46}\\A_{51}&A_{52}&A_{53}&A_{54}&A_{55}&A_{56}\\A_{61}&A_{62}&A_{63}&A_{64}&A_{65}&A_{66}\end{bmatrix}} $

Die Vektoren und Matrizen in Voigt'scher Notation können addiert, subtrahiert und mit einem Skalar multipliziert werden. Beim Matrizenprodukt in Voigt'scher Notation muss eine Diagonalmatrix

$ I=\mathrm {diag} (1,1,1,2,2,2) $

mit den Einträgen $ I_{uv}=\mathbf {S} _{u}:\mathbf {S} _{v} $ zwischengeschaltet werden:

$ \mathbf {A} :\mathbf {B} =[\mathbf {A} ]^{\top }I[\mathbf {B} ]=A_{1}B_{1}+A_{2}B_{2}+A_{3}B_{3}+2A_{4}B_{4}+2A_{5}B_{5}+2A_{6}B_{6} $
$ \left[{\stackrel {4}{\mathbf {A} }}:\mathbf {T} \right]=\left[{\stackrel {4}{\mathbf {A} }}\right]I[\mathbf {T} ] $
$ \left[{\stackrel {4}{\mathbf {A} }}:{\stackrel {4}{\mathbf {B} }}\right]=\left[{\stackrel {4}{\mathbf {A} }}\right]I\left[{\stackrel {4}{\mathbf {B} }}\right] $

Darin steht [x] für die Voigt-Notation von x.

Einzelnachweise

  1. P. B. Denton, S. J. Parke, T. Tao, X. Zhang: Eigenvectors from Eigenvalues. (PDF) 10. August 2019, S. 1–3, abgerufen am 29. November 2019 (Lua-Fehler in Modul:Multilingual, Zeile 149: attempt to index field 'data' (a nil value)).
  2. J. Hanson: Rotations in three, four, and five dimensions. Bei: arxiv.org. S. 4f.

Literatur

  • Holm Altenbach: Kontinuumsmechanik. Einführung in die materialunabhängigen und materialabhängigen Gleichungen. 2. Auflage. Springer Vieweg, Berlin u. a. 2012, ISBN 978-3-642-24118-5.
  • Philippe Ciarlet: Mathematical Elasticity. Band 1: Three-Dimensional Elasticity. North-Holland, Amsterdam 1988, ISBN 0-444-70259-8.
  • Wolfgang Ehlers: Ergänzung zu den Vorlesungen Technische Mechanik und Höhere Mechanik. Vektor- und Tensorrechnung, Eine Einführung. 2015 (uni-stuttgart.de [PDF; abgerufen am 3. September 2020]).
  • Ralf Greve: Kontinuumsmechanik. Ein Grundkurs für Ingenieure und Physiker. Springer, Berlin u. a. 2003, ISBN 3-540-00760-1.