Differential Equations II: Linear ODE Systems
"Differential Equations II: Linear ODE Systems" refers to the study of systems of ordinary differential equations (ODEs) where the equations are linear in the unknown functions and their derivatives. This topic typically covers methods for solving such systems, including matrix exponentials, eigenvalues and eigenvectors, and phase plane analysis. It is a continuation of introductory differential equations, focusing on the theory, solutions, and applications of linear ODE systems in various scientific fields.
Differential Equations II: Linear ODE Systems
"Differential Equations II: Linear ODE Systems" refers to the study of systems of ordinary differential equations (ODEs) where the equations are linear in the unknown functions and their derivatives. This topic typically covers methods for solving such systems, including matrix exponentials, eigenvalues and eigenvectors, and phase plane analysis. It is a continuation of introductory differential equations, focusing on the theory, solutions, and applications of linear ODE systems in various scientific fields.
💡 Key Takeaways
- Represent linear ODE systems as x' = A x and identify the system matrix A.
- Solve constant-coefficient systems using the matrix exponential e^{At} and the fundamental matrix.
- Diagonalize A using eigenvalues and eigenvectors (or use Jordan form when A is not diagonalizable) to obtain the general solution.
- Analyze stability and long-term behavior from the eigenvalues and initial data, and visualize with phase portraits.
❓ Frequently Asked Questions
What is a linear ODE system and how is it written?
A linear system of first-order ODEs for a vector x(t) can be written as x'(t) = A x(t), where x(t) ∈ R^n and A is an n×n constant matrix. Initial values specify x(0).
How do you solve x' = A x using eigenvalues and eigenvectors?
Compute the eigenpairs (λ, v) from det(A − λI) = 0. Each eigenpair gives a solution x(t) = e^{λ t} v. If A has n independent eigenvectors, the general solution is a linear combination of these: ∑ c_i e^{λ_i t} v_i. If A is not diagonalizable, use generalized eigenvectors and Jordan blocks, yielding terms like t^k e^{λ t} v.
What is the matrix exponential and how does it solve linear systems?
The solution with x(0) = x0 is x(t) = e^{A t} x0. The matrix exponential e^{A t} = ∑_{k=0}^∞ (A t)^k / k!, and it satisfies d/dt e^{A t} = A e^{A t} with e^{0} = I.
When is diagonalization helpful, and what if A isn’t diagonalizable?
If A = P D P^{-1} with D diagonal, then x(t) = P e^{D t} P^{-1} x0, which reduces to decoupled scalar equations. If A isn’t diagonalizable, use Jordan form A = P J P^{-1} and x(t) = P e^{J t} P^{-1} x0, which may introduce terms like t e^{λ t}.
How can you assess the long-term behavior of solutions from the eigenvalues?
The real parts of eigenvalues determine stability. If all Re(λ) < 0, x(t) → 0 as t → ∞ (stable). If any Re(λ) > 0, some solutions grow without bound (unstable). If some Re(λ) = 0, marginal stability requires further analysis.