Frequency Domain Curve-Fitting FDCF

Frequency domain curve-fitting is useful technique to fit a TFM very close into the observed FRF data. FDCF is a two-step procedure, which includes the model structure selection and model parameter optimization.

The first step is to parameterize the TFM in some special forms. Two forms of parameterizing are introduced in the following: the matrix fraction (MF) parameterization and the polynomial matrix (PM) parameterization. Parameterizing is always a critical step in the identification due to the reason that it will generally lead to quite different parameter optimization algorithms, in addition to this, it results in model properties.

This tutorial clarify that for the MF form, the parameters can be optimized by the help of linear least squares (LLS) solutions. As for the PM parameterization, you have alternative to some nonlinear techniques; specifically. The PM parameterization offers more flexibility in the context that it allows the designer to specify certain properties of the identified model (e.g., allows the designer to fix zeros in any input/output channels).