LiesThirdTheorem - Maple Help
For the best experience, we recommend viewing online help using Google Chrome or Mozilla Firefox.

Online Help

All Products    Maple    MapleSim


GroupActions[LiesThirdTheorem] - find a Lie algebra of pointwise independent vector fields with prescribed structure equations (solvable algebras only)

Calling Sequences

     LiesThirdTheorem(Alg, M, option)

     LiesThirdTheorem(A, M)

Parameters

     Alg       - a Maple name or string, the name of an initialized Lie algebra 𝔤

     M         - a Maple name or string, the name of an initialized manifold with the same dimension as that of 𝔤

     option    - with output = "forms" the dual 1-forms (Maurer-Cartan forms) are returned

     A         - a  list of square matrices, defining a matrix Lie algebra

 

Description

Examples

Description

• 

Let g be an n−dimensional Lie algebra with structure constants C. Then Lie's Third Theorem (see, for example, Flanders, page 108) asserts that there is, at least locally, a Lie algebra of n pointwise independent vector fields Γ on an n-dimensional manifold M with structure constants C.

• 

The command LiesThirdTheorem(Alg, M) produces a globally defined Lie algebra of vector fields Γ in the special case that 𝔤 is solvable. More general cases will be handled in subsequent versions of the DifferentialGeometry package.

• 

The command LiesThirdTheorem(A, M) produces a globally defined matrix of 1-forms (Maurer-Cartan forms) in the special case that the list of matrices A defines a solvable Lie algebra.

• 

The command LiesThirdTheorem is part of the DifferentialGeometry:-GroupActions package. It can be used in the form LiesThirdTheorem(...) only after executing the commands with(DifferentialGeometry) and with(GroupActions), but can always be used by executing DifferentialGeometry:-GroupActions:-LiesThirdTheorem(...).

Examples

> 

with⁡DifferentialGeometry:with⁡LieAlgebras:with⁡GroupActions:with⁡Library:

 

Example 1.

We obtain a Lie algebra from the DifferentialGeometry library using the Retrieve command and initialize it.

> 

L≔Retrieve⁡Winternitz,1,4,4,Alg1

L:=e1,e4=e1,e2,e4=e2+e1,e3,e4=e3+e2

(2.1)
> 

DGsetup⁡L:

 

We define a manifold M of dimension 4 (the same dimension as the Lie algebra).

Alg1 > 

DGsetup⁡x,y,z,w,M1

frame name: M1

(2.2)
M1 > 

Γ1≔LiesThirdTheorem⁡Alg1,M1

Γ1:=D_x,D_y,D_z,x+y⁢D_x+y+z⁢D_y+z⁢D_z+D_w

(2.3)
M1 > 

Ω1≔LiesThirdTheorem⁡Alg1,M1,output=forms

Ω1:=dx−x+y⁢dw,dy−y+z⁢dw,−dw⁢z+dz,dw

(2.4)

 

We calculate the structure equations for the Lie algebra of vector fields Gamma1 and check that these structure equations coincide with those for Alg1.

M1 > 

LieAlgebraData⁡Γ1,Alg1a

e1,e4=e1,e2,e4=e2+e1,e3,e4=e3+e2

(2.5)

 

Example 2.

We re-work the previous example in a more complicated basis. In this basis the adjoint representation is not upper triangular, in which case LiesThirdTheorem first calls the program SolvableRepresentation to find a basis for the algebra in which the adjoint representation is upper triangular. (Remark: It is almost always useful, when working with solvable algebras, to transform to a basis where the adjoint representation is upper triangular.)

M1 > 

L2≔LieAlgebraData⁡e4,e2−e4,e3,e1+e3,Alg2

L2:=e1,e2=−e4+e3−e2−e1,e1,e3=−e3−e2−e1,e1,e4=−e4−e2−e1,e2,e3=e3+e2+e1,e2,e4=e4+e2+e1

(2.6)
Alg2 > 

DGsetup⁡L2

Lie algebra: Alg2

(2.7)
Alg2 > 

Γ2≔LiesThirdTheorem⁡Alg2,M1

Γ2:=x−y⁢D_x+y+z⁢D_y+z⁢D_z+D_w,−x−y⁢D_x−−1+y+z⁢D_y−z⁢D_z−D_w,D_z,−D_x+D_z

(2.8)
M1 > 

LieAlgebraData⁡Γ2,Alg2a

e1,e2=−e4+e3−e2−e1,e1,e3=−e3−e2−e1,e1,e4=−e4−e2−e1,e2,e3=e3+e2+e1,e2,e4=e4+e2+e1

(2.9)

 

Example 3.

Here is an example where one of the adjoint matrices has complex eigenvalues. The Lie algebra contains parameters p and b.

M1 > 

L3≔Retrieve⁡Winternitz,1,5,25,Alg3

L3:=e1,e5=2⁢_p⁢e1,e2,e3=e1,e2,e5=_p⁢e2+e3,e3,e5=_p⁢e3−e2,e4,e5=_b⁢e4

(2.10)
M1 > 

DGsetup⁡L3:

Alg3 > 

Adjoint⁡e5

Alg3 > 

DGsetup⁡x,y,z,u,v,M3

frame name: M3

(2.11)
M3 > 

Γ3≔LiesThirdTheorem⁡Alg3,M3

Γ3:=D_x,−D_x⁢z+D_y,D_z,D_u,12⁢z2−12⁢y2+2⁢_p⁢x⁢D_x+_p⁢y−z⁢D_y+_p⁢z+y⁢D_z+_b⁢u⁢D_u+D_v

(2.12)
M3 > 

LieAlgebraData⁡Γ3,Alg3a

e1,e5=2⁢_p⁢e1,e2,e3=e1,e2,e5=_p⁢e2+e3,e3,e5=_p⁢e3−e2,e4,e5=_b⁢e4

(2.13)

 

Example 4.

We calculate the Maurer-Cartan matrix Ω of 1-forms for a solvable matrix algebra, namely the matrices defining the adjoint representation for Alg1 from Example 1.

> 

A≔Adjoint⁡Alg1

> 

MaurerCartan≔LiesThirdTheorem⁡A,M1

 

Note that the elements of this matrix   coincide with the appropriate linear combinations of the forms in the list Ω1 from Example 1.

Alg1 > 

MaurerCartan1,4,Ω11&plusΩ12

dx+dy−2⁢y+z+x⁢dw,dx+dy−2⁢y+z+x⁢dw

(2.14)
Alg1 > 

MaurerCartan2,4,Ω12&plusΩ13

dy+dz−2⁢z+y⁢dw,dy+dz−2⁢z+y⁢dw

(2.15)
Alg1 > 

MaurerCartan3,4,Ω13

−dw⁢z+dz,−dw⁢z+dz

(2.16)
Alg1 > 

MaurerCartan1,1,−1&multΩ14

−dw,−dw

(2.17)

See Also

DifferentialGeometry

GroupActions

Library

LieAlgebras

Adjoint

LieAlgebraData

Representation

Retrieve

SolvableRepresentation