Application of modified Arnoldi algorithm to passive macromodeling of MEMS

Woon Ket Wong,Wei Wang
DOI: https://doi.org/10.1166/jnn.2009.c123
Abstract:The demand for accurately simulating dynamical responses of complex MEMS and NEMS systems leads to intensive studies in reduced-order modeling methods. We apply a modified Block Arnoldi algorithm to significantly reduce the run time and usage of computer resource for such calculations, while preserving essential properties. The 2n x 2n matrix in the computation is replaced by a n x n matrix and the FLOP count is reduced from (56n(3) - 216n(2) + 22n) / 3 to (7n(3) - 54n(2) + 11n) / 3. The CPU run time for a resonator example of n = 39 is reduced from 0.091 second to 0.080 second. For a butterfly gyroscope example with a larger matrix size, n = 17361, the CPU time is reduced from 4343 seconds to 1528 seconds, achieving 65% improvement.
What problem does this paper attempt to address?