Find the normal form of a conic

  • Thread starter Thread starter sara_87
  • Start date Start date
  • Tags Tags
    Form Normal
Click For Summary

Homework Help Overview

The discussion revolves around finding the normal form of a quadratic equation representing a conic section, specifically the equation (x^2) - 3xy + (y^2) = 1. Participants are exploring the transformation of this equation into matrix form and discussing the implications of the eigenvalues and eigenvectors in identifying the type of conic section represented.

Discussion Character

  • Exploratory, Conceptual clarification, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • Participants discuss the initial steps of rewriting the quadratic form as a matrix and the process of diagonalization. Questions arise regarding the treatment of the constant on the right-hand side and the implications of the eigenvalues on the type of conic section. There is also exploration of finding eigenvectors corresponding to the eigenvalues.

Discussion Status

Participants are actively engaging with the problem, sharing insights about eigenvalues and eigenvectors, and questioning each other's reasoning. Some guidance has been provided regarding the properties of symmetric matrices and the relationship between eigenvalues and conic types. There is an ongoing exploration of the implications of the derived eigenvectors.

Contextual Notes

Participants are working under the constraints of homework rules, focusing on understanding the mathematical concepts without providing direct solutions. There is a noted confusion regarding the eigenvalues and eigenvectors, which is being clarified through discussion.

sara_87
Messages
748
Reaction score
0

Homework Statement



(a) find the normal form of the following quadratic form
(b) describe the conic section

(x^2)-3xy+(y^2)=1

i know we must put it as a matrix first but i don't know what to do about the 1 on the right hand side.
Thank you.
 
Physics news on Phys.org
Ignore it to start with. There are several ways to "normalize" a quadratic but since you specifically mention "matrices", I suspect what you are doing is writing the quadratic as
\left(\begin{array}{cc} x & y\end{array}\right)\left(\begin{array}{cc}1 & -3/2 \\ -3/2 & 1\end{array}\right)\left(\begin{array}{c}x \\ y\end{array}\right)= 1

Since the middle matrix is symmetric, it can be diagonalized. Find its eigenvalues and eigenvectors to diagonalize it and then in the x', y' coordinates having the eigenvectors as axes, the matrix will be diagonal and
\left(\begin{array}{cc} x' & y'\end{array}\right)\left(\begin{array}{cc}\lambda_1 & 0 \\ 0 & \lambda_2\end{array}\right)\left(\begin{array}{c}x' \\ y'\end{array}\right)= 1
What kind of conic section that is will, of course, depend upon whether \lambda_1 and \lambda_2 are positive, negative, or 0.
 
this helps a lot. thanx.
but in your matrix, u wrote -3/2 twice...but surely one of them must be positive.
so if e-values are -ve, it's hyperbola, if +ve, ellipse and if zero, circle
is that right?
 
the e-values are: sqrt13/2 and -sqrt13/2
 
i'm struuggling to find the eigenvectors
 
sara_87 said:
this helps a lot. thanx.
but in your matrix, u wrote -3/2 twice...but surely one of them must be positive.
so if e-values are -ve, it's hyperbola, if +ve, ellipse and if zero, circle
is that right?
No. A matrix giving a quadratic form is always symmetric.
\left(\begin{array}{cc} x & y\end{array}\right)\left(\begin{array}{cc}1 & -3/2 \\ -3/2 & 1\end{array}\right)\left(\begin{array}{c}x \\ y\end{array}\right)= 1
= \left(\begin{array}{cc}x- 3/2y \\-3/2 x+ y\end{array}\right)\left(\begin{array}{c}x \\ y\end{array}\right)
= x^2- 3xy+ y^2
Because it is always symmetric, a quadratic form always has real eigenvalues.

If both eigenvalues have the same sign, then you have an ellipse (or its "special forms", circle, point or empty set). If the eigenvalues have different signs then a hyperbola, if one is 0, then parabola.

The eigenvalue equation is
\left|\begin{array}{cc}1-\lambda &amp; -3/2 \\ -3/2 &amp; 1-\lambda\end{array}\right|= (1- \lambda)^2- 9/4= 0[/itex]<br /> so the eigenvalues are much simpler than you got!
 
Last edited by a moderator:
so eigenvalues are (sqrt5)/2 and -(sqrt5)/2
?
 
What in the world are you doing?

What are the roots of (\lambda- 1)^2= 9/4?
 
Oh my god
hahahahahahahaha
wt was i doing? I'm so ashamed of myself! i expanded wrong.
so the eigenvalues are 5/2 and -1/2
so now i cud find the eigen vectors!
 
  • #10
i am struggling to find the eigen vectors.
 
  • #11
The fact that 5/2 is an eigenvalue means that, for any eigenvector <x, y>
\left(\begin{array}{cc}1 &amp; -3/2 \\ -3/2 &amp; 1\end{array}\right)\left(\begin{array}{c}x \\ y\end{array}\right)= \left(\begin{array}{c}(5/2)x \\ (5/2)y \end{array}\right)
which, equating the two rows gives two equations: x- (3/2)y= (5/2)x and -(3/2)x+ y= (5/2)y. Because 5/2 is an eigenvector those two equations are NOT independent: they both reduce to y= -x. Any x,y satisfying that gives an eigenvector corresponding to eigenvalue 5/2: in particular, taking x= 1, y= -1, <1, -1> is an eigenvector corresponding to eigenvalue 5/2.

Can you do the same thing for eigenvalue -1/2?
 
  • #12
oh ok. i worked on it and got the same result for -1/2.
so now what will i do with the =1?
 
  • #13
No, two different eigenvalues cannot have the same eigenvectors. In fact, eigenvectors, corresponding to different eigenvalues, of a symmetric matrix must be perpendicular.

Please show me exactly how you "got the same result for -1/2".
 
  • #14
oh sorry i made a mistake, the eigenvector is not the same, it is <1, 1>
 
  • #15
And, of course, the vector <1, 1> points along the line y= x or y- x= 0 while the vector <1, -1> points along the line y= -x or y+ x= 0. Let u= y- x, v= y+ x (so that x= ? , y= ?) and write the equation in terms of the variables u and v by replacing x and y with their expressions in terms of u and v.
 
  • #16
so x=(v-u)/2
and y=(v+u)/2
 
  • #17
Yes, now substitute those in your equation: x2- 3xy+ y2= 1.
 
  • #18
this gives:
(5u^2)-(v^2)=4
 

Similar threads

  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 11 ·
Replies
11
Views
8K
Replies
4
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 6 ·
Replies
6
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 3 ·
Replies
3
Views
2K
  • · Replies 7 ·
Replies
7
Views
1K
Replies
24
Views
3K
Replies
5
Views
3K