15

How can I do vector calculations without telling Mathematica the vector entries?

I have very many arbitrary linear combinations in $\mathbb{R}^3$ which I want to perform some general calculations on (scalar and vector products) and want to use Mathematica to do this (especially for simplifying stuff like very long equations with scalar prodcuts of vector products).

So, I don't want to write for all my vectors stuff like v1={a1,b1,c1} ... vN={aN,bN,cN} and so on, but just want to say v1 ... vN are vectors.

How is this possible?

Foo Bar
  • 335
  • 2
  • 8

2 Answers2

21

Assuming that we have three-dimensional real vectors :

$Assumptions = (u | v | w) ∈ Vectors[3, Reals];

we can use e.g. various tensor functions (new in ver. 9) e.g. TensorReduce to reduce (simplify) a tensor expression, e.g.

TensorReduce[ v.v + w.w - (v + w).(v + w) ]
TensorReduce[u \[Cross] (v \[Cross] w) ]
-2 v.w
-w u.v + v u.w

We can perform more interesting reductions, let's show e.g. the Jacobi identity:

TensorReduce[ u \[Cross] (v \[Cross] w) + v \[Cross] (w \[Cross] u) + 
              w \[Cross] (u \[Cross] v) ]
0

or write it in a traditional form:

Defer[   u \[Cross] (v \[Cross] w) + v \[Cross] (w \[Cross] u)
       + w \[Cross] (u \[Cross] v)] == 
TensorReduce[   u \[Cross] (v \[Cross] w) + v \[Cross] (w \[Cross] u)
              + w \[Cross] (u \[Cross] v) ] // TraditionalForm

enter image description here

Another common identity

TensorExpand[ (u \[Cross] v) \[Cross] (u \[Cross] w) ]
TensorExpand[ (u \[Cross] v).(u \[Cross] v) ]
u u \[Cross] v . w
-(u.v)^2 + u.u  v.v

Take a look at new differential operators:

Curl[ Curl[ f[x, y, z], {x, y, z}], {x, y, z}] == Laplacian[ f[x, y, z], {x, y, z}]
True
Sjoerd C. de Vries
  • 65,815
  • 14
  • 188
  • 323
Artes
  • 57,212
  • 12
  • 157
  • 245
  • {1, 2, 3} \[Element] Vectors[3, Reals] evaluates to True, yet v + {1, 2, 3} evaluates to {1+v,2+v,3+v}. Any suggestions on how to properly suppress this Listable attribute of Plus where it isn't appropriate? – RRas Aug 08 '19 at 16:23
6

@Szabolcs is right, use Symbolic Tensors. But in that link it may be a bit confusing to find what you want. There are good examples on 3D vector operations. Read:

For example, proving an identity:

a\[Cross](b\[Cross](c\[Cross]d)) == b a.(c\[Cross]d) - (a.b) c\[Cross]d // TensorExpand

True

Or expanding something very long:

((a\[Cross]b).c)^4 // TensorExpand

(a.c)^4 (b.b)^2 - 4 a.b (a.c)^3 b.b b.c + 4 (a.b)^2 (a.c)^2 (b.c)^2 + 2 a.a (a.c)^2 b.b (b.c)^2 - 4 a.a a.b a.c (b.c)^3 + (a.a)^2 (b.c)^4 + 2 (a.b)^2 (a.c)^2 b.b c.c - 2 a.a (a.c)^2 (b.b)^2 c.c - 4 (a.b)^3 a.c b.c c.c + 4 a.a a.b a.c b.b b.c c.c + 2 a.a (a.b)^2 (b.c)^2 c.c - 2 (a.a)^2 b.b (b.c)^2 c.c + (a.b)^4 (c.c)^2 - 2 a.a (a.b)^2 b.b (c.c)^2 + (a.a)^2 (b.b)^2 (c.c)^2

Vitaliy Kaurov
  • 73,078
  • 9
  • 204
  • 355