Tensor in D-dimensional space crosswise with 2 vectors

  • Context: Graduate 
  • Thread starter Thread starter olgerm
  • Start date Start date
  • Tags Tags
    Space Tensor Vectors
Click For Summary

Discussion Overview

The discussion revolves around the concept of constructing a tensor in D-dimensional Euclidean space that is orthogonal to two given vectors. Participants explore the properties and conditions of such a tensor, as well as the implications of its construction for specific vector relationships.

Discussion Character

  • Exploratory
  • Technical explanation
  • Conceptual clarification
  • Debate/contested

Main Points Raised

  • One participant introduces two vectors, ##\vec {V}## and ##\vec {X}##, and inquires about a tensor that is orthogonal to both, suggesting it might be parallel to a (D-2)-dimensional area.
  • Another participant questions the clarity of the request and proposes that a rank 2 tensor could be considered, presenting two conditions that such a tensor might satisfy involving the vectors.
  • A different participant expresses uncertainty about how to describe their needs in terms of tensors, instead framing the problem in terms of four vectors and the relationship between them, particularly focusing on the vector ##\vec F## being orthogonal to ##\vec V_2## and parallel to a plane defined by ##\vec X## and ##\vec V_1##.
  • One participant seeks confirmation on whether a tensor that meets their requirements exists, indicating a desire to understand the limitations of their inquiry.

Areas of Agreement / Disagreement

Participants do not reach a consensus on the existence or construction of the desired tensor. Multiple interpretations and approaches are presented, indicating ongoing uncertainty and exploration of the topic.

Contextual Notes

The discussion includes various assumptions about the properties of tensors and the relationships between vectors, which remain unresolved. The notation and specific definitions of the tensors and vectors involved are not uniformly agreed upon.

Who May Find This Useful

This discussion may be of interest to individuals exploring tensor mathematics, particularly in the context of vector relationships in higher-dimensional spaces.

olgerm
Gold Member
Messages
535
Reaction score
35
I have 2 vectors ##\vec {V}=(v_1,v_2,v_3,v_4...v_D)## and ##\vec {X}=(x_1,x_2,x_3,x_4...x_D)## in D-dimensional euclidean space.
I want a tensor ,which is crosswise with both of them.

I think that the tensor is parallel with (D-2)-dimensional area, am I right?
I do not know a lot about tensors ,so in answer please also explain me the notation that you used.
 
Physics news on Phys.org
It's not exactly clear what you want. Do you mean a rank 2 tensor, which we could think of as a matrix? In that case, we can have two possible conditions:

1. ##\sum_j M_{ij} v_j =0##, or
2. ##\sum_j v_j M_{ji}=0##.

You might also want both of these to be satisfied. These conditions are independent unless ##M## is symmetric, ##M_{ij}=M_{ji}##.

Either way, an analog of the Gram-Schmidt process allows you to start with any tensor ##M## and construct a new tensor ##M'## that satisfies one or both of these conditions. For instance, as long as ##|V|\neq 0##,
$$ M'_{ij} = M_{ij} - \sum_k M_{ik} v_k \frac{v_j}{|V|^2}$$
will satisfy ##\sum_j M'_{ij} v_j = 0##. The other combinations should be easy to work out.

Hopefully this component notation is familiar, or at least obvious to you.
 
  • Like
Likes   Reactions: olgerm
fzero said:
Do you mean a rank 2 tensor, which we could think of as a matrix?
I can´t explain ,what I mean ,in terms of tensor ,but I´ll try to explain it in terms of what I want it for:
There are 4 vectors ##\vec V_1## ,##\vec V_2## ,##\vec X## and ##\vec F##.
##F⊥V_2##
F is parallel with plane ,which is also parallel with vectors ##\vec X## and ##\vec V_1##

I know only value of vectors ##\vec X## and ##\vec V_1##. My friend knows only value of vector ##\vec V_2##. I want a tensor that I could give to my friend so he could calculate value of ##\vec F## without knowing vectors ##\vec X## and ##\vec V_1## separately.
 
Nb: by value I meant the vector itself not length of vector.

I there any such tensor at all ,I wanted? If there is not, it would be useful to learn that there is not.
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 5 ·
Replies
5
Views
4K
  • · Replies 4 ·
Replies
4
Views
3K
  • · Replies 25 ·
Replies
25
Views
4K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 11 ·
Replies
11
Views
2K
Replies
6
Views
1K
  • · Replies 40 ·
2
Replies
40
Views
4K
  • · Replies 57 ·
2
Replies
57
Views
8K
  • · Replies 5 ·
Replies
5
Views
2K