Prove the following identities using index notation

  • Thread starter ponjavic
  • Start date
  • Tags
    Tensor
In summary: However, I cannot find the equation for (A^t). I'm guessing that it has something to do with the fact that (A^t) is a skew symmetric tensor, but I'm not sure how to prove it.
  • #1
ponjavic
225
0
I hope no one minds but I'm going to keep all my questions and solutions here and then write it down for the report =)

Homework Statement


1 Prove the following identities using index notation:
a) For vectors, u, v, w,
(u cross v) dot w = u dot (v cross w) XsolvedX

b)
For two second order tensors, show that trace(AB) = trace(BA) using index notation

c)
For two second order tensors, show that trace(AB) = trace(BA) using index notation
(AB)^t = B^t A^t

2 Consider a tensor V and a vector w related as follows
V_kj = e_ijk*w_i

a) Write out V as a 3 x 3 matrix

b) Show, using index notation, that V is a skew symmetric tensor (V = -V^t)

c) Show, using index notation or otherwise, that for any vector, v,
Vv = w x v

d) Show, using index notation, that the vector w is written in terms of V by,
w_i = 1/2 * e_ijk * V_kj

The vector w is known as the vector dual of the skew symmetric tensor V

Note: For part (d) you may make use of one or the both identities:
e_ijk * e_ipq = delta_jp * delta_kq - delta_kp * delta_jq
e_ijk * e_jkr = e_ijk * e_rjk = 2delta_ir

Homework Equations


a)
(u dot v) = uivj
(u cross v) = eijkujvk
(eijk is the permutation tensor)

b)
trace(A) = Aii

The Attempt at a Solution


(u cross v) dot w= eijkujvk * wi =
eijkwiujvk =
ejkiuivjwk =
(as eijk = ejki) number of permutations are even:
eijkvjwk * ui=
u dot (v cross w)

solved

b)
trace(AB) = trace(ABij) = trace(AikBkj) = AikBki = BkiAik = trace(BkiAjk) =
trace(BAij) = trace(BA)

c)
A=Aik, A^t = Aki
B=Bkj, B^t = Bjk
(AB)^t = (AikBkj)^t .. not quite sure how the transpose operation would work on a multiplication

2
a)
V_kj =
w1, w1, w1
w2, w2, w2
w3, w3, w3

No idea what to do with e_ijk...

b)
For A to be skew symmetric, A^t = -A

V_kj^t = ... Do not know how to transpose e_ijk * w_i
possibly: e_ikj * w_k
Then permutation k <-> j:
-e_ijk * w_k = -V_kj

Is this allowed?
 
Last edited:
Physics news on Phys.org
  • #2
a) your cyclic permuttation of ijk is correct because e_ijk is invarinat ulnder cyclic permutation.

b) recall (AB)_ij=AikBkj. Now the trace sets i=j and sums.
 
  • #3
edit: check first post
 
Last edited:
  • #4
Comon people I'm really struggling with this, no one done tensors?
 
  • #5
ponjavic said:
Comon people I'm really struggling with this, no one done tensors?

Sorry, I forgot that we were at your beck and call :rolleyes:
 
  • #6
cristo said:
Sorry, I forgot that we were at your beck and call :rolleyes:
Helpful, I hope you can do tensors and aren't just wasting my time
 
  • #7
For (c) you know how to write a product, you know how to write a transpose, and so you should be able to write the transpose of that product.

The parts you've answered from 2 look correct.
 

FAQ: Prove the following identities using index notation

1. What is a tensor problem?

A tensor problem refers to a mathematical or computational problem that involves tensors, which are mathematical objects used to represent physical quantities such as force, velocity, and stress in a multidimensional space.

2. How are tensors used in scientific research?

Tensors are used in a wide range of scientific fields, including physics, engineering, and computer science, to model and understand complex systems and phenomena. They are also commonly used in machine learning and data analysis.

3. What are some common applications of tensors?

Tensors have many practical applications, such as image and signal processing, data compression, and fluid dynamics simulations. They are also used in materials science, quantum mechanics, and general relativity.

4. What are the challenges in solving tensor problems?

One of the main challenges in solving tensor problems is the high dimensionality of tensor data, which can lead to computational difficulties and require specialized algorithms. Another challenge is the interpretation of results, as tensors can represent complex relationships between multiple variables.

5. Are there any resources for learning about tensors and tensor problems?

Yes, there are many resources available for learning about tensors and how to solve tensor problems. These include textbooks, online courses, and research papers. Additionally, there are software packages and libraries specifically designed for working with tensors, such as TensorFlow and PyTorch.

Back
Top