Can All Functions Be Parameterized?

  • Context: Undergrad 
  • Thread starter Thread starter darkar
  • Start date Start date
  • Tags Tags
    Function
Click For Summary

Discussion Overview

The discussion centers around the parameterization of functions, particularly in the context of scalar and vector functions. Participants explore methods for parameterizing curves defined by level curves and discuss the existence of parameterizations for various types of functions.

Discussion Character

  • Exploratory
  • Technical explanation
  • Debate/contested

Main Points Raised

  • One participant inquires about specific methods for parameterizing functions of scalar and vector types, referencing a particular example from a website.
  • Another participant suggests that no universal method exists for parameterization, but provides examples of parameterizing level curves by solving for one variable in terms of another.
  • A counterpoint is raised regarding the surjectivity of the cosh function, indicating that it does not cover the entire real line, thus challenging the completeness of the proposed parameterization.
  • A further contribution emphasizes that there are infinitely many parameterizations for any given curve, illustrating this with a piecewise parameterization of a path that includes corners.
  • A participant expresses curiosity about the existence of functions that cannot be parameterized, indicating a desire for further exploration of the topic.

Areas of Agreement / Disagreement

Participants generally agree that there is no single method for parameterizing all curves, and multiple competing views on specific parameterization techniques are present. The discussion remains unresolved regarding the existence of functions that cannot be parameterized.

Contextual Notes

Some assumptions about the smoothness of curves and the conditions under which parameterizations are valid are not fully explored. The discussion also touches on the limitations of certain mathematical functions in covering specific ranges.

darkar
Messages
187
Reaction score
0
Is there a specific method to parameterised a function of scalar and vector?

I was reading through this website but then, can't understand how it read taht parameterised steps.

Any guide?

(http://www.math.umn.edu/~nykamp/m2374/readings/stokesex/index.html
under double check example)

Thanks
 
Mathematics news on Phys.org
I don't think there is a method that works in all cases. But when the curve is specified as a level curve f(x,y)=0, you can try to solve for y. In that case, just use x as the parameter.

Exemple: Consider the level curve C={(x,y): y-x²-2x=0. We can solve of y like so: y=x²+2x. In that case, the parametrized curve is r(t)=(t²+2t,t), t [itex]\in \mathbb{R}[/itex].

In other cases, if the relation between x and y is something familiar you can try to exploit this as illustrated in the following exemple:

Exemple: Consider the level curve C={(x,y): (x/a)²-(y/b)²=1}. This resembles the identity cosh²(t)-sinh²(t)=1. So set x(t)/a=cosh(t) and y(t)/b=sinh(t), i.e. let r(t)=(acosh(t),bsinh(t)) t [itex]\in \mathbb{R}[/itex] parametrize the level curve.

N.B. You can convince yourself that this parametrization covers the whole curve because given any value of x, there is a corresponding value of t for which acosh(t)=x and similarly for y. [cosh(t) is surjective on the x-axis and sinh(t) is surjective on the y axis]. See http://en.wikipedia.org/wiki/Image:Sinh_cosh_tanh.svg
 
No wait, cosh(t) is not surjective on the real line, it only covers (1, infty). So r(t) would only cover part of C.
 
As quasar987 said, there is no one method of parameterizing a curve. In fact, there exist an infinite number of different parameterizations for any curve.

For the example given, because the path (From the origin, (0, 0, 0) along the z-axis to (0, 0 1), then along the quarter circle to (0, 1, 0), then along the y-axis to (0, 0, 0) again) is not "smooth" (there are corners at (0,0,0), (0, 1, 0) and (0, 0, 1)), you would break it into three pieces.

First the line from (0, 0, 0) to (0, 0, 1). At any point on that line, x=y= 0
An obvious parameterization is to use z itself as parameter: x= 0, y= 0, z= t, with [itex]0\le t\le 1[/itex].
Second, the quarter circle from (0, 0, 1) to (0, 1, 0). At every point on that circle x= 0 and y2+ z2= 1. A "standard" parameterization for a circle is to use sine and cosine: x= 0, y= sin(t), z= cos(t). Then x= 0 always while y2+ z2= sin2(t)+ cos2(t)= 1 for all t. Of course, it is z= cos(t) and not y because when t= 0, z= cos(0)= 1 and y= sin(0)= 0 as required. When t= [itex]\pi/2[/itex], z= cos([itex]\pi/2[/itex])= 0 and y= sin([itex]\pi/2[/itex])= 1 so [itex]0\le t\le \pi/2[/itex].

Finally, the line from (0, 1, 0) to (0, 0, 0). Obviously x= z= 0 at every point. We could use y itself as parameter: x= 0, y= t, z= 0 with t going from 1 to 0. Another possibility is x= 0, y= 1- t, z= 0 with t going from 0 to 1.
 
Oh rite, i see.
Thanks for helps!

Edit: oh, btw, is there any functions that can't be parameterized?
 
Last edited:

Similar threads

  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 11 ·
Replies
11
Views
3K
  • · Replies 23 ·
Replies
23
Views
9K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 6 ·
Replies
6
Views
3K
  • · Replies 17 ·
Replies
17
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 1 ·
Replies
1
Views
4K
  • · Replies 3 ·
Replies
3
Views
4K
  • · Replies 2 ·
Replies
2
Views
3K