nwidPEM function
Identify a full network or a single target module using prediction error minimization (PEM), based on the implementation in the System Identification toolbox. The network model is transformed into a state space model to support optimization with pem
.
Syntax
model = nwidPEM(data,model,orders,options)
performs identification based on the provided data, initialized network model, model orders and option set. The identified predictor model is returned.
Input arguments
data
nwdata
objectNetwork data object. The input excitation, input node and output node signals (as referred to by their labels) in the
Rset
,Dset
andYset
of the predictor model must be present in the data.model
nwpredmodel
objectnwmodel
objectInitialized network model object. Use an
nwmodel
object to identify the full network, and annwpredmodel
object to identify a single target module. For each element, whether or not it is fixed, the number of delays, the orders of the numerators and denominators, parameter values of fixed elements and parameter initializations can be set using theG
,T
andH
fields (PredictorMap
objects).H
must specify a square matrix and is enforced to be monic by the algorithm.orders
structure array Model orders. Can be used to override properties of the initialized network model object. Use a structure array with optional fields
G
,T
andH
, which should each be structure arrays themselves with the optional fieldsnpnum
,npden
andd
(to set the analogousPredictorMap
properties). Use an empty structure array to use the initialized network model object without modifications.options
nwidPEMOptions
objectOption set for
nwidPEM
function. Can be constructed usingoptions = nwidPEMOptions
for the default option set oroptions = nwidPEMOptions(Name=Value,...)
to set specified options.
Output arguments
model
nwpredmodel
objectnwmodel
objectEstimated network model object. This is an
nwmodel
object after full network identification, and annwpredmodel
object after single module identification. Thenpnum
andnpden
properties of theG
,T
andH
fields are populated with the estimated parameter values. Thee
properties ofG
,T
andH
contain the fixed and estimated transfer functions of the predictor model connections.