Piecewise approach for matrix-interpolation-based parametric Model Order Reduction

In the design process of mechanical systems dynamic analysis is carried out in time domain or in frequency domain which implies solving the equation of motion several times. Usually the systems depend on a set of parameters which influence their responses. Thus the design process includes numerical simulations using a full-scale finite element model for each set of parameters which is computational demanding and time consuming, e.g. in the scope of frequency response. One approach to tackle this problem is by using parametric model order reduction (pMOR) methods. In this contribution a piecewise approach is presented for those pMOR methods based on interpolation in matrix manifolds and reduced order model (ROM) databases. The piecewise approach decreases the computational time and the stored data during the parametric-based simulation. A numerical example illustrates the suitability of this method in special for large frequency bands. The respective results using the piecewise pMOR approach is compared with the solution obtained using the corresponding full-scale model.      «

In the design process of mechanical systems dynamic analysis is carried out in time domain or in frequency domain which implies solving the equation of motion several times. Usually the systems depend on a set of parameters which influence their responses. Thus the design process includes numerical simulations using a full-scale finite element model for each set of parameters which is computational demanding and time consuming, e.g. in the scope of frequency response. One approach to tackle thi...     »