Applying Vector Fitting for Measurement-based Multiple-Input Multiple-Output Model Identification of a Grid-Forming Converter

This paper presents an approach for applying Vector
Fitting (VF) to identify a small-signal state-space model of
a power electronic converter represented as a multiple-input
multiple-output (MIMO) system. A workflow for the proposed
model identification procedure is presented, with focus on
aspects related to the processing of data from measurements
and on the configuration of the VF algorithm. An example of
how a small-signal state-space model can be identified from
laboratory measurements on a 50 kW grid-forming converter
is presented to demonstrate how VF is suitable for identifying
black-box MIMO models of power converters. The identification
procedure is applied to the full matrix of transfer functions
between the input and output combinations obtained from
the laboratory measurements. The accuracy of the resulting
state-space model is verified in the frequency-domain and by
time-domain simulations. Finally, limitations of the presented
approach are discussed.
Index Terms—power electronic systems, model identification,
state-space models, MIMO systems, laboratory experiments