91Ó°ÊÓ

A nonhomogeneous linear system refers to a system of differential equations where there is an external input or forcing term involved. Unlike homogeneous systems that have terms entirely dependent on the state variables, nonhomogeneous systems include constant matrices or vector terms that introduce changes to the state that are independent of the current state itself.
This is represented in the format:

• \[ \mathbf{y}' = A\mathbf{y} + \mathbf{g}_0 \]

where \( \mathbf{y} \) is the vector of state variables, \( A \) is the matrix of coefficients, and \( \mathbf{g}_0 \) provides the nonhomogeneous component. In these systems, finding the equilibrium point can be more challenging, as it includes both solving for underlying homogeneous parts and accounting for the shifts introduced by \( \mathbf{g}_0 \). The typical strategy involves a transformation to reframe it as a homogeneous system, facilitating easier analysis.

Expressing a system of differential equations in matrix form helps in managing and solving the system effectively. It consolidates information into matrices, making complex calculations more manageable using linear algebra techniques.
For example, consider a system:

• \[ x' = -x + 2 \]
• \[ y' = 2y - 4 \]

This can be expressed in the matrix equation:

• \[ \mathbf{y}' = A\mathbf{y} + \mathbf{g}_0 \]

where \(A = \begin{bmatrix} -1 & 0 \ 0 & 2 \end{bmatrix},\mathbf{y} = \begin{bmatrix} x \ y \end{bmatrix},and \mathbf{g}_0 = \begin{bmatrix} 2 \ -4 \end{bmatrix}.\)This transformation facilitates the use of tools like eigenvalues and eigenvectors to analyze system behavior, stability, and responses to perturbations.

Eigenvalues are intrinsic values that offer insight into the stability and behavior of linear systems. They help to determine how solutions to the differential equations behave over time. For a matrix \( A \) in matrix form, eigenvalues \( \lambda \) are solutions to the characteristic equation:

• \[ \text{det}(A - \lambda I) = 0 \]

where \( I \) is the identity matrix.
In the considered system, the eigenvalues for the matrix \( A = \begin{bmatrix} -1 & 0 \ 0 & 2 \end{bmatrix} \) are \( \lambda_1 = -1 \) and \( \lambda_2 = 2 \).

• Real eigenvalues: Indicate straightforward growth or decay along the corresponding direction.
• Complex eigenvalues would suggest oscillatory behavior.

The specific combination of real and distinct eigenvalues in our example indicates a saddle point, thus offering significant insight into the system's dynamics.

Stability classification determines the robustness of an equilibrium point when subjected to small deviations. Based on the signs of eigenvalues, we can classify stability as:

• Stable: If all eigenvalues are negative, any deviation from the equilibrium decays over time, returning to the equilibrium.
• Unstable: If at least one eigenvalue is positive, deviations from the equilibrium grow over time.
• Saddle Point: If eigenvalues have mixed signs, deviations in certain directions grow while others decay.

In our system, the eigenvalues \( \lambda_1 = -1 \) and \( \lambda_2 = 2 \) signify a saddle point. This classification conveys that the equilibrium is unstable, as deviations will diverge along the positive eigenvalue’s direction. Identifying stability classification is crucial for understanding potential system outcomes and long-term behavior.