A linear differential system has vector form .
Three nonnegative amounts convert in the directed cycle at positive first-order rates . The linear system of ordinary differential equations conserves . Its characteristic polynomial is . The two nonzero eigenvalues have negative real part, so every trajectory converges to the unique steady state with that total:
This is also a column-convention continuous-time Markov chain after normalizing by : every component loses to one successor and receives from one predecessor. A complex-conjugate pair produces damped oscillations about equilibrium even though the amounts remain nonnegative. A repeated nonzero eigenvalue is defective and requires a Jordan chain.
When the displayed equality holds, the two nonzero eigenvalues of the three-state cyclic conversion model coincide at . The positive cyclic rates force each eigenspace to have dimension at most one: if its coordinate vanished, the first eigenvector equation would force , and then the third would force . Thus the repeated eigenvalue has only one eigenvector direction and the matrix is not diagonalizable. A length-two Jordan chain produces both and transients. The rates give an explicit example with .
For , substituting an exponential particular solution reduces the differential equation to . The displayed solution exists whenever is not an eigenvalue of . Resonance can require polynomial factors in time. The general solution adds , expressed using the matrix exponential.
An eigenvector of with eigenvalue gives a homogeneous solution for . If is diagonalizable, these powers form a fundamental system. A constant vector forcing has a particular solution when is not an eigenvalue. At a resonant eigenvalue or a nontrivial Jordan block, logarithmic factors in can occur, equivalently by changing time to and solving a constant-coefficient linear system.
For with constant matrix , an eigenvector with eigenvalue gives . If has an eigenbasis, every solution is the linear combination , with constants fixed by the initial condition. For a real matrix, complex conjugate eigenpairs combine into real oscillatory solutions.
For , the eigenvectors and have eigenvalues . Data give
Substitution verifies both equations and the data. A defective matrix needs generalized eigenvectors and polynomial-exponential Jordan chains, as in the defective repeated eigenvalue in a cyclic conversion model; eigenvectors alone then do not span all solutions.[ref]
For a homogeneous system , its state transition matrix maps the state at time to the state at time :
For continuous coefficients on an interval, uniqueness gives and . If is any fundamental matrix, then , independent of that choice. A forced system has the variation of constants formula
For constant , . Generally it is not , since matrices at different times need not commute. For piecewise constant on and on , chronological propagation gives . With and , this product is , while has both diagonal entries . This exhibits the importance of time order.[ref]
The real planar system , becomes the complex scalar equation under . For constant , : the modulus scales exponentially and the argument rotates at angular speed . For , the radius is preserved, giving uniform rotation.
A general real linear system , instead has
These formulas follow by substituting , . Only the rotation-dilation case , makes . Thus encoding two real components as one complex variable does not make every real planar system complex-linear.

Articles by others on the same topic (0)

There are currently no matching articles.