Fast Passivity Assessment of Admittance Parameter Macromodels by a Half-Size Singularity Test Matrix

The passivity of delay-less state-space models is often assessed via the eigenvalues of a Hamiltonian matrix which is established from the model. Any purely imaginary eigenvalue corresponds to a boundary frequency for passivity violations. In this paper we investigate an alternative (non-Hamiltonian) matrix for passivity assessment, that was recently introduced. The new matrix is only half the size of the Hamiltonian, and so the eigenvalue computation is about eight times faster. In addition, the new test matrix has certain numerical advantages.