One of steps in the design of microwave filters is numerical tuning using full-wave simulators. Typically, it is a time-consuming process as it uses advanced computational methods, e.g. the finite-element method (FEM) and it usually requires multiple optimization steps before the specification goals are met. FEM involves solving a large sparse system of equations at many frequency points and therefore its computational cost is high. One of the ideas to speed up the numerical optimization is parametrized model-order reduction (PMOR). The key point in model order reduction, is that the original large sparse system of FE equations is replaced with a small and dense one, that can be solved at many frequency points with substantially smaller computational effort. PMOR yields in the parameter dependent reduced-order model which might be reused in subsequent optimization steps.
Authors
Additional information
- Category
- Aktywność konferencyjna
- Type
- publikacja w wydawnictwie zbiorowym recenzowanym (także w materiałach konferencyjnych)
- Language
- angielski
- Publication year
- 2020
