Options set for
opt = procestOptions
opt = procestOptions(Name,Value)
Specify optional pairs of arguments as
the argument name and
Value is the corresponding value.
Name-value arguments must appear after other arguments, but the order of the
pairs does not matter.
Before R2021a, use commas to separate each name and value, and enclose
Name in quotes.
InitialCondition — Handling of initial conditions
'auto' (default) |
Handling of initial conditions during estimation, specified as one of the following values:
'zero'— The initial condition is set to zero.
'estimate'— The initial condition is treated as an independent estimation parameter.
'backcast'— The initial condition is estimated using the best least squares fit.
'auto'— The software chooses the method to handle initial condition based on the estimation data.
DisturbanceModel — Handling of additive noise
'estimate' (default) |
Handling of additive noise (H) during estimation for the model
e is white noise, u is the input and y is the output.
H(s) is stored in the
of the numerator and denominator of
DisturbanceModel is specified as one of
the following values:
'none'— H is fixed to one.
'estimate'— H is treated as an estimation parameter. The software uses the value of the
NoiseTFproperty as the initial guess.
'ARMA1'— The software estimates H as a first-order ARMA model
'ARMA2'— The software estimates H as a second-order ARMA model
'fixed'— The software fixes the value of the
NoiseTFproperty of the
idprocmodel as the value of H.
A noise model cannot be estimated using frequency domain data.
InputInterSample — Input-channel intersample behavior
Input-channel intersample behavior for transformations between discrete time and continuous time, specified as
The definitions of the three behavior values are as follows:
'zoh'— Zero-order hold maintains a piecewise-constant input signal between samples.
'foh'— First-order hold maintains a piecewise-linear input signal between samples.
'bl'— Band-limited behavior specifies that the continuous-time input signal has zero power above the Nyquist frequency.
iddata objects have a similar property,
data.InterSample, that contains the same behavior value options.
InputInterSample value is
the estimation data is in an
software uses the
data.InterSample value. When the estimation data
is instead contained in a timetable or a matrix pair, with the
option, the software uses
The software applies the same option value to all channels and all experiments.
InputOffset — Removal of offset from time-domain input data
'auto' (default) |
'estimate' | vector | matrix | object |
Removal of offset from time-domain input data during estimation, specified as one of the following values:
'estimate'— The software treats the input offsets as an estimation parameter.
'auto'— The software chooses the method to handle input offsets based on the estimation data and the model structure. The estimation either assumes zero input offset or estimates the input offset.
For example, the software estimates the input offset for a model that contains an integrator.
A column vector of length Nu, where Nu is the number of inputs.
to specify no offsets.
In case of multi-experiment data, specify
InputOffsetas a Nu-by-Ne matrix. Nu is the number of inputs, and Ne is the number of experiments.
Each entry specified by
InputOffsetis subtracted from the corresponding input data.
A parameter object, constructed using
param.Continuous, that imposes constraints on how the software estimates the input offset.
For example, create a parameter object for a 2-input model estimation. Specify the first input offset as fixed to zero and the second input offset as an estimation parameter.
opt = procestOptions; u0 = param.Continuous('u0',[0;NaN]); u0.Free(1) = false; opt.Inputoffset = u0;
Advanced — Additional advanced options
Advanced is a structure with the following
ErrorThreshold— Specifies when to adjust the weight of large errors from quadratic to linear.
Errors larger than
ErrorThresholdtimes the estimated standard deviation have a linear weight in the loss function. The standard deviation is estimated robustly as the median of the absolute deviations from the median of the prediction errors, divided by
0.7. For more information on robust norm choices, see section 15.2 of .
ErrorThreshold = 0disables robustification and leads to a purely quadratic loss function. When estimating with frequency-domain data, the software sets
ErrorThresholdto zero. For time-domain data that contains outliers, try setting
MaxSize— Specifies the maximum number of elements in a segment when input-output data is split into segments.
MaxSizemust be a positive integer.
StabilityThreshold— Specifies thresholds for stability tests.
StabilityThresholdis a structure with the following fields:
s— Specifies the location of the right-most pole to test the stability of continuous-time models. A model is considered stable when its right-most pole is to the left of
z— Specifies the maximum distance of all poles from the origin to test stability of discrete-time models. A model is considered stable if all poles are within the distance
zfrom the origin.
AutoInitThreshold— Specifies when to automatically estimate the initial condition.
The initial condition is estimated when
ymeas is the measured output.
yp,z is the predicted output of a model estimated using zero initial states.
yp,e is the predicted output of a model estimated using estimated initial states.
Create Default Option Set for Process Model Estimation
opt = procestOptions;
Specify Options for Process Model Estimation
Create an option set for
'simulation' and turning on the
opt = procestOptions('Focus','simulation','Display','on');
Alternatively, use dot notation to set the values of
opt = procestOptions; opt.Focus = 'simulation'; opt.Display = 'on';
 Ljung, L. System Identification: Theory for the User. Upper Saddle River, NJ: Prentice-Hall PTR, 1999.
 Wills, Adrian, B. Ninness, and S. Gibson. “On Gradient-Based Search for Multivariable System Estimates”. Proceedings of the 16th IFAC World Congress, Prague, Czech Republic, July 3–8, 2005. Oxford, UK: Elsevier Ltd., 2005.
Version HistoryIntroduced in R2012a
InputInterSample option allows intersample behavior specification for continuous models estimated from timetables or matrices.
iddata objects contain an
InterSample property that
describes the behavior of the signal between sample points. The
InputInterSample option implements a version of that property in
procestOptions so that intersample behavior can be specified also when
estimation data is stored in timetables or matrices.
R2018a: Renaming of Estimation and Analysis Options
The names of some estimation and analysis options were changed in R2018a. Prior names still work. For details, see the R2018a release note Renaming of Estimation and Analysis Options.