Stable spiral but math says saddle

  • Context: MHB 
  • Thread starter Thread starter Dustinsfl
  • Start date Start date
  • Tags Tags
    Spiral Stable
Click For Summary
SUMMARY

The discussion centers on the analysis of fixed points in a dynamical system defined by the equations $$\dot{x} = x(3 - 2x - y)$$ and $$\dot{y} = y(-5 + 5x)$$. The fixed points identified are $(0,0)$, $\left(\frac{3}{2},0\right)$, and $\left(1,1\right)$. The Jacobian matrix at each fixed point indicates that all three points are saddles, despite the expectation of a stable spiral at $(1,1)$. The issue arises from a sign error in the Jacobian matrix, specifically in the [2,1] entry.

PREREQUISITES
  • Understanding of dynamical systems and fixed points
  • Knowledge of Jacobian matrices and their role in stability analysis
  • Familiarity with eigenvalues and eigenvectors
  • Basic proficiency in mathematical notation and analysis
NEXT STEPS
  • Review the calculation of Jacobian matrices in dynamical systems
  • Study the implications of eigenvalues on the stability of fixed points
  • Learn about the classification of fixed points in nonlinear systems
  • Examine common pitfalls in stability analysis, particularly sign errors
USEFUL FOR

Mathematicians, physicists, and engineers working with dynamical systems, particularly those analyzing stability and fixed points in nonlinear equations.

Dustinsfl
Messages
2,217
Reaction score
5
At my fixed point of (1,1), I should have a stable spiral but I keep getting analytically that it is a saddle.

$$
\begin{cases} \dot{x} = x(3 - 2x - y)\\
\dot{y} = y(-5 + 5x)
\end{cases}
$$
The fixed points are $(0,0)$, $\left(\frac{3}{2},0\right)$, and $\left(1,1\right)$.
Also, the Jacobian is
$$
\mathcal{J} = \begin{pmatrix}
3 - 4x - y & -x\\
-5y & -5 + 5x
\end{pmatrix}.
$$
At $(0,0)$, we have $\mathcal{J} = \begin{pmatrix}3 & 0\\0 & -5\end{pmatrix}$; that is, $\text{tr} = -2$, $\det = -15$, and the discriminant is greater than 0.
Therefore, we have an saddle at $(0,0)$.
For $\lambda = 3$, the eigenvector is $e_1$, so when $\lambda = 5$, the eigenvector is $e_2$.At $\left(\frac{3}{2},0\right)$, we have $\mathcal{J} = \begin{pmatrix}-3 & -\frac{3}{2}\\0 & \frac{5}{2}\end{pmatrix}$; that is, $\text{tr} = -\frac{1}{2}$, $\det = -\frac{15}{2}$, and the discriminant is greater than 0.
Therefore, we have a saddle at $\left(\frac{3}{2},0\right)$.
For $\lambda = -3$, the eigenvector is $e_1$, so when $\lambda = \frac{5}{2}$, the eigenvector is $\begin{pmatrix}-\frac{3}{11}\\1\end{pmatrix}$.At $\left(1,1\right)$, we have $\mathcal{J} = \begin{pmatrix}-2 & -1\\-5 & 0\end{pmatrix}$; that is, $\text{tr} = -2$, $\det = -5$, and the discriminant is greater than 0.
Therefore, we have a saddle at $\left(1,1\right)$.
For $\lambda = i\sqrt{5}$, the eigenvector is $\begin{pmatrix}\frac{i}{\sqrt{5}}\\1\end{pmatrix}$, so when $\lambda = -i\sqrt{5}$, the eigenvector is $\begin{pmatrix}-\frac{i}{\sqrt{5}}\\1\end{pmatrix}$.

View attachment 389
It is obviously spiral. I have re-checked and done the problem three times and nothing is working out. What is going wrong?
 

Attachments

  • phasebsecond.jpg
    phasebsecond.jpg
    33.8 KB · Views: 77
Physics news on Phys.org
Check the [2,1] entry of your Jacobian matrix. I see a sign problem.
 

Similar threads

  • · Replies 9 ·
Replies
9
Views
3K
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 8 ·
Replies
8
Views
2K
  • · Replies 6 ·
Replies
6
Views
1K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 1 ·
Replies
1
Views
2K
Replies
9
Views
2K
  • · Replies 35 ·
2
Replies
35
Views
7K
  • · Replies 3 ·
Replies
3
Views
2K