Is the System Stiff? Analyzing Eigenvalues of A

  • MHB
  • Thread starter evinda
  • Start date
  • Tags
    System
In summary, the discussion revolves around the stiffness of a specific system of differential equations. The system in question has two equations and two initial values, and it is shown that the eigenvalues are -1.001 and 0.001. The definition of a stiff system is mentioned, and it is noted that the system may not meet this definition due to having only one negative eigenvalue. However, it is suggested that the magnitude of the negative eigenvalue may still indicate stiff-like behavior in the system. Ultimately, the determination of whether the system is considered stiff may depend on the specific criteria and definitions used by the professor or the assistant.
  • #1
evinda
Gold Member
MHB
3,836
0
Hello! (Wave)

Let the problem of initial values

$\left\{\begin{matrix}
y_1'(t)=-0.5y_1(t)+0.501 y_2(t), t \in [0,10^3]\\
y_2'(t)=0.501y_1(t)-0.5y_2(t) \\
y_1(0)=1.1 \\
y_2(0)-0.9
\end{matrix}\right. \\ $

The question is if the system is stiff.$$A=\bigl(\begin{smallmatrix}
-0.5 & 0.501\\
0.501 & -0.5
\end{smallmatrix}\bigr)$$

$$y'=Ay$$

where $y'=\binom{y_1'}{y_2'}, \ y=\binom{y_1}{y_2}$.

The eigenvalues are $\lambda_1=-1.001, \lambda_2=0.001$.According the assistant of the professor, since it holds that $|\lambda_1|>> |\lambda_2|$ we deduce that the system is stiff.
But I found the following in my book:>Stiff systems are systems of the form $x'=Ax$ for ehich there are eigenvalues $\lambda_{\mu}$ and $\lambda_{\nu}$ of $A$, with negative real part, such that $|Re{\lambda_{\mu}}|>>|Re{\lambda_{\nu}}|$.So since we found only one negative and one positive eigenvalue, don't we conclude that the system isn't stiff? Or am I wrong? :confused:
 
Mathematics news on Phys.org
  • #2


Hello there! As a fellow scientist, let me offer my perspective on this problem.

Based on the definition of stiff systems that you found in your book, it seems like the system in question may not be considered stiff. However, it is important to note that the concept of stiffness is not always well-defined and can vary depending on the context and the problem at hand.

In this case, it may be helpful to also consider the values of the eigenvalues in addition to their signs. Even though there is only one negative eigenvalue, the magnitude of that eigenvalue is much larger than the positive eigenvalue. This indicates that the system may exhibit behavior similar to that of a stiff system, even if it may not technically meet the strict definition.

Ultimately, whether or not the system is considered stiff may depend on the specific criteria and definitions used by the professor or the assistant. It may be helpful to discuss this further with them to gain a better understanding of their reasoning and perspective on the problem.

I hope this helps and good luck with your analysis!
 

FAQ: Is the System Stiff? Analyzing Eigenvalues of A

How do I determine if a system is stiff?

To determine if a system is stiff, you can analyze the eigenvalues of the system's matrix A. If the eigenvalues have a large difference in magnitude, then the system is likely stiff. Additionally, if the eigenvalues are complex and have a small imaginary part, the system may also be stiff.

What is the significance of eigenvalues in determining stiffness?

Eigenvalues represent the behavior of a system over time. When analyzing the eigenvalues of a system's matrix A, a large difference in magnitude indicates that the system is difficult to solve numerically and may require special methods to accurately simulate its behavior.

Can a system be both stiff and non-stiff?

Yes, a system can exhibit both stiff and non-stiff behavior depending on the initial conditions and parameters of the system. It is possible for a system to have both stiff and non-stiff regions, which may require different numerical methods for accurate simulation.

How do I handle a stiff system in my simulations?

There are several methods for handling stiff systems in simulations. One approach is to use implicit methods, such as the backward differentiation formula, which can better handle larger differences in eigenvalues. Another approach is to use adaptive step-size control, which adjusts the time step based on the behavior of the system.

Can I prevent a system from becoming stiff?

In some cases, it may be possible to prevent a system from becoming stiff by carefully choosing the parameters and initial conditions of the system. However, in many cases, stiffness is an inherent property of the system and cannot be avoided. In these cases, it is important to choose appropriate numerical methods for accurately simulating the system's behavior.

Similar threads

Replies
1
Views
1K
Replies
2
Views
2K
Replies
4
Views
2K
Replies
6
Views
2K
Replies
9
Views
2K
Replies
1
Views
2K
Back
Top