Passive Reduced-Order Macromodeling Algorithm for Microelectromechanical Systems

In this paper, we present a passive reduced-order macromodeling algorithm for second-order dynamic systems of linear microelectromechanical systems (MEMS) devices. The proposed reduction algorithm is based on congruent transformations. The system equations of MEMS devices, given by finite-element methods (FEMs), are converted to state-space forms that are compatible with passive Krylov subspace methods. To achieve this, a modified matrix equation is proposed for second-order MEMS dynamics. In addition, the generalized procedure is provided for first-order heat transfer problems and second-order structure dynamic problems to ensure that the discretized FEM system satisfies all the necessary conditions to guarantee the passivity of the reduced-order system. Finally, numerical examples are provided to demonstrate the validity of the proposed passive reduction technique. $\hfill$ [2006-0224]