Documentation

This is machine translation

Translated by Microsoft
Mouseover text to see original. Click the button below to return to the English verison of the page.

Note: This page has been translated by MathWorks. Please click here
To view all translated materals including this page, select Japan from the country navigator on the bottom of this page.

System Identification Toolbox Functions

Alphabetical List By Category

Data Preparation

Represent Data

iddataTime- or frequency-domain data
idfrdFrequency-response data or model
idinputGenerate input signals
simSimulate response of identified model
sizeQuery output/input/array dimensions of input–output model and number of frequencies of FRD model
midprefsSpecify location for file containing System Identification app startup information
simOptionsOption set for sim

Select Data for Estimation

fselectSelect frequency points or range in FRD model
getexpSpecific experiments from multiple-experiment data set
merge (iddata)Merge data sets into iddata object
fcatConcatenate FRD models along frequency dimension

Analyze Data

bodeBode plot of frequency response, or magnitude and phase data
bodemagBode magnitude response of LTI models
plotPlot input-output data
adviceAnalysis and recommendations for data or estimated linear models
delayestEstimate time delay (dead time) from data
isrealDetermine whether model parameters or data values are real
realdataDetermine whether iddata is based on real-valued signals
feedbackIdentify possible feedback data
pexcitLevel of excitation of input signals
impulseestNonparameteric impulse response estimation
etfeEstimate empirical transfer functions and periodograms
spaEstimate frequency response with fixed frequency resolution using spectral analysis
spafdrEstimate frequency response and spectrum using spectral analysis with frequency-dependent resolution
iddataPlotOptionsOption set for iddata/plot

Preprocess Data

detrendSubtract offset or trend from data signals
retrendAdd offsets or trends to data signals
diffDifference signals in iddata objects
idfiltFilter data using user-defined passbands, general filters, or Butterworth filters
misdataReconstruct missing input and output data
nkshiftShift data sequences
idresampResample time-domain data by decimation or interpolation
resampleResample time-domain data by decimation or interpolation (requires Signal Processing Toolbox software)
getTrendData offset and trend information
chgFreqUnitChange frequency units of frequency-response data model
fdelDelete specified data from frequency response data (FRD) models
TrendInfoOffset and linear trend slope values for detrending data

Transform Data

fftTransform iddata object to frequency domain data
ifftTransform iddata objects from frequency to time domain
etfeEstimate empirical transfer functions and periodograms
spaEstimate frequency response with fixed frequency resolution using spectral analysis
spafdrEstimate frequency response and spectrum using spectral analysis with frequency-dependent resolution

Linear Model Identification

Process Models

procestEstimate process model using time or frequency data
idprocContinuous-time process model with identifiable parameters
pemPrediction error estimate for linear and nonlinear model
idparCreate parameter for initial states and input level estimation
delayestEstimate time delay (dead time) from data
initSet or randomize initial parameter values
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getparObtain attributes such as values and bounds of linear model parameters
setparSet attributes such as values and bounds of linear model parameters
procestOptionsOptions set for procest

Input-Output Polynomial Models

arxEstimate parameters of ARX or AR model using least squares
armaxEstimate parameters of ARMAX model using time-domain data
bjEstimate Box-Jenkins polynomial model using time domain data
iv4ARX model estimation using four-stage instrumental variable method
ivxARX model estimation using instrumental variable method with arbitrary instruments
oeEstimate Output-Error polynomial model using time or frequency domain data
polyestEstimate polynomial model using time- or frequency-domain data
idpolyPolynomial model with identifiable parameters
pemPrediction error estimate for linear and nonlinear model
arxstrucCompute loss functions for single-output ARX models
ivstrucCompute loss functions for sets of ARX model structures using instrumental variable method
selstrucSelect model order for single-output ARX models
strucGenerate model-order combinations for single-output ARX model estimation
arxRegulDetermine regularization constants for ARX model estimation
delayestEstimate time delay (dead time) from data
initSet or randomize initial parameter values
polydataAccess polynomial coefficients and uncertainties of identified model
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getparObtain attributes such as values and bounds of linear model parameters
setparSet attributes such as values and bounds of linear model parameters
setPolyFormatSpecify format for B and F polynomials of multi-input polynomial model
armaxOptionsOption set for armax
arxOptionsOption set for arx
arxRegulOptionsOption set for arxRegul
bjOptionsOption set for bj
iv4OptionsOption set for iv4
oeOptionsOption set for oe
polyestOptionsOption set for polyest

State-Space Models

ssestEstimate state-space model using time or frequency domain data
ssregestEstimate state-space model by reduction of regularized ARX model
n4sidEstimate state-space model using subspace method
idssState-space model with identifiable parameters
pemPrediction error estimate for linear and nonlinear model
delayestEstimate time delay (dead time) from data
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getparObtain attributes such as values and bounds of linear model parameters
setparSet attributes such as values and bounds of linear model parameters
ssformQuick configuration of state-space model structure
initSet or randomize initial parameter values
idparCreate parameter for initial states and input level estimation
idssdataState-space data of identified system
findstatesEstimate initial states of model
ssestOptionsOption set for ssest
ssregestOptionsOption set for ssregest
n4sidOptionsOption set for n4sid
findstatesOptionsOption set for findstates

Transfer Function Models

tfestTransfer function estimation
idtfTransfer function model with identifiable parameters
pemPrediction error estimate for linear and nonlinear model
delayestEstimate time delay (dead time) from data
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getparObtain attributes such as values and bounds of linear model parameters
setparSet attributes such as values and bounds of linear model parameters
tfdataAccess transfer function data
initSet or randomize initial parameter values
tfestOptionsOption set for tfest

Linear Grey-Box Models

greyestLinear grey-box model estimation
idgreyLinear ODE (grey-box model) with identifiable parameters
pemPrediction error estimate for linear and nonlinear model
findstatesEstimate initial states of model
initSet or randomize initial parameter values
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getparObtain attributes such as values and bounds of linear model parameters
setparSet attributes such as values and bounds of linear model parameters
findstatesOptionsOption set for findstates
greyestOptionsOption set for greyest

Frequency-Response Models

etfeEstimate empirical transfer functions and periodograms
spaEstimate frequency response with fixed frequency resolution using spectral analysis
spafdrEstimate frequency response and spectrum using spectral analysis with frequency-dependent resolution
idfrdFrequency-response data or model
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
bodeBode plot of frequency response, or magnitude and phase data
bodemagBode magnitude response of LTI models
freqrespFrequency response over grid
chgFreqUnitChange frequency units of frequency-response data model

Correlation Models

craEstimate impulse response using prewhitened-based correlation analysis
impulseestNonparameteric impulse response estimation
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
impulseestOptionsOptions set for impulseest

Nonlinear Model Identification

Nonlinear ARX Models

idnlarxNonlinear ARX model
nlarxEstimate parameters of nonlinear ARX model
nlarxOptionsOption set for nlarx
isnlarxDetect nonlinearity in estimation data
initSet or randomize initial parameter values
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
polyregPowers and products of standard regressors
customregCustom regressor for nonlinear ARX models
addregAdd custom regressors to nonlinear ARX model
getregRegressor expressions and numerical values in nonlinear ARX model
customnetCustom nonlinearity estimator for nonlinear ARX and Hammerstein-Wiener models
linearClass representing linear nonlinearity estimator for nonlinear ARX models
neuralnetClass representing neural network nonlinearity estimator for nonlinear ARX models
treepartitionClass representing binary-tree nonlinearity estimator for nonlinear ARX models
wavenetCreate a wavelet network nonlinearity estimator object
sigmoidnetClass representing sigmoid network nonlinearity estimator for nonlinear ARX and Hammerstein-Wiener models
evaluateValue of nonlinearity estimator at given input
simSimulate response of identified model
simOptionsOption set for sim
predictPredict K-step ahead model output
predictOptionsOption set for predict
compareCompare model output and measured output
compareOptionsOption set for compare
forecastForecast identified model output
forecastOptionsOption set for forecast
plotPlot nonlinearity of nonlinear ARX model
evaluateValue of nonlinearity estimator at given input
getDelayInfoGet input/output delay information for idnlarx model structure
findopCompute operating point for Nonlinear ARX model
findopOptionsOption set for findop
operspecConstruct operating point specification object for idnlarx model
linearizeLinearize nonlinear ARX model
linappLinear approximation of nonlinear ARX and Hammerstein-Wiener models for given input

Hammerstein-Wiener Models

idnlhwHammerstein-Wiener model
nlhwEstimate Hammerstein-Wiener model
nlhwOptionsOption set for nlhw
initSet or randomize initial parameter values
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
customnetCustom nonlinearity estimator for nonlinear ARX and Hammerstein-Wiener models
deadzoneCreate a dead-zone nonlinearity estimator object
poly1dClass representing single-variable polynomial nonlinear estimator for Hammerstein-Wiener models
pwlinearCreate a piecewise-linear nonlinearity estimator object
saturationCreate a saturation nonlinearity estimator object
sigmoidnetClass representing sigmoid network nonlinearity estimator for nonlinear ARX and Hammerstein-Wiener models
unitgainSpecify absence of nonlinearities for specific input or output channels in Hammerstein-Wiener models
wavenetCreate a wavelet network nonlinearity estimator object
evaluateValue of nonlinearity estimator at given input
simSimulate response of identified model
simOptionsOption set for sim
compareCompare model output and measured output
compareOptionsOption set for compare
plotPlot input and output nonlinearity, and linear responses of Hammerstein-Wiener model
evaluateValue of nonlinearity estimator at given input
findopCompute operating point for Hammerstein-Wiener model
findopOptionsOption set for findop
operspecConstruct operating point specification object for idnlhw model
linearizeLinearize Hammerstein-Wiener model
linappLinear approximation of nonlinear ARX and Hammerstein-Wiener models for given input

Nonlinear Grey-Box Models

nlgreyestEstimate nonlinear grey-box model parameters
idnlgreyNonlinear grey-box model
pemPrediction error estimate for linear and nonlinear model
findstatesEstimate initial states of model
initSet or randomize initial parameter values
getinitValues of idnlgrey model initial states
setinitSet initial states of idnlgrey model object
getparParameter values and properties of idnlgrey model parameters
setparSet initial parameter values of idnlgrey model object
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
simSimulate response of identified model
nlgreyestOptionsOption set for nlgreyest
findstatesOptionsOption set for findstates
simOptionsOption set for sim

Grey-Box Model Estimation

greyestLinear grey-box model estimation
nlgreyestEstimate nonlinear grey-box model parameters
idgreyLinear ODE (grey-box model) with identifiable parameters
idnlgreyNonlinear grey-box model
pemPrediction error estimate for linear and nonlinear model
findstatesEstimate initial states of model
initSet or randomize initial parameter values
getinitValues of idnlgrey model initial states
setinitSet initial states of idnlgrey model object
getparParameter values and properties of idnlgrey model parameters
setparSet initial parameter values of idnlgrey model object
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
simSimulate response of identified model
greyestOptionsOption set for greyest
nlgreyestOptionsOption set for nlgreyest
findstatesOptionsOption set for findstates
simOptionsOption set for sim

Model Validation

Compare Output with Measured Data

compareCompare model output and measured output
goodnessOfFitGoodness of fit between test and reference data
findstatesEstimate initial states of model
idparCreate parameter for initial states and input level estimation
compareOptionsOption set for compare
findstatesOptionsOption set for findstates

Residual Analysis

residCompute and test residuals
pePrediction error for identified model
fpeAkaike’s Final Prediction Error for estimated model
aicAkaike’s Information Criterion for estimated model
residOptionsOption set for resid
peOptionsOption set for pe

Uncertainty Analysis

presentDisplay model information, including estimated uncertainty
simsdSimulate linear models with uncertainty using Monte Carlo method
freqrespFrequency response over grid
rsampleRandom sampling of linear identified systems
showConfidenceDisplay confidence regions on response plots for identified models
getcovParameter covariance of identified model
setcovSet parameter covariance data in identified model
translatecovTranslate parameter covariance across model transformation operations
stepStep response plot of dynamic system; step response data
stepplotPlot step response and return plot handle
impulseImpulse response plot of dynamic system; impulse response data
bodeBode plot of frequency response, or magnitude and phase data
bodemagBode magnitude response of LTI models
nyquistNyquist plot of frequency response
nyquistplotNyquist plot with additional plot customization options
iopzmapPlot pole-zero map for I/O pairs of model
iopzplotPlot pole-zero map for I/O pairs and return plot handle
tfdataAccess transfer function data
zpkdataAccess zero-pole-gain data
simsdOptionsOption set for simsd

Model Analysis

Continuous- and Discrete-Time Conversions

c2dConvert model from continuous to discrete time
d2cConvert model from discrete to continuous time
d2dResample discrete-time model
translatecovTranslate parameter covariance across model transformation operations
c2dOptionsCreate option set for continuous- to discrete-time conversions
d2cOptionsCreate option set for discrete- to continuous-time conversions
d2dOptionsCreate option set for discrete-time resampling

Model Type and Other Transformations

idfrdFrequency-response data or model
idpolyPolynomial model with identifiable parameters
idtfTransfer function model with identifiable parameters
idssState-space model with identifiable parameters
canonState-space canonical realization
balredModel order reduction
noisecnvTransform identified linear model with noise channels to model with measured channels only
translatecovTranslate parameter covariance across model transformation operations
mergeMerge estimated models
appendGroup models by appending their inputs and outputs
noise2measNoise component of model
absorbDelayReplace time delays by poles at z = 0 or phase shift
chgTimeUnitChange time units of dynamic system
chgFreqUnitChange frequency units of frequency-response data model
fdelDelete specified data from frequency response data (FRD) models
stackBuild model array by stacking models or model arrays along array dimensions
ss2ssState coordinate transformation for state-space model

Linearization of Nonlinear Models

linappLinear approximation of nonlinear ARX and Hammerstein-Wiener models for given input
idnlarx/findopCompute operating point for Nonlinear ARX model
idnlarx/linearizeLinearize nonlinear ARX model
idnlhw/findopCompute operating point for Hammerstein-Wiener model
idnlhw/linearizeLinearize Hammerstein-Wiener model
findopOptionsOption set for findop

Data Extraction

polydataAccess polynomial coefficients and uncertainties of identified model
ssdataAccess state-space model data
idssdataState-space data of identified system
tfdataAccess transfer function data
zpkdataAccess zero-pole-gain data
frdataAccess data for frequency response data (FRD) object
freqrespFrequency response over grid
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
getcovParameter covariance of identified model
setcovSet parameter covariance data in identified model
getAccess model property values
setSet or modify model properties
nparamsNumber of model parameters
ndimsQuery number of dimensions of dynamic system model or model array
orderQuery model order
poleCompute poles of dynamic system
zeroZeros and gain of SISO dynamic system
sizeQuery output/input/array dimensions of input–output model and number of frequencies of FRD model
dampNatural frequency and damping ratio
dcgainLow-frequency (DC) gain of LTI system
bandwidthFrequency response bandwidth

Simulation and Prediction

simSimulate response of identified model
simOptionsOption set for sim
simsdSimulate linear models with uncertainty using Monte Carlo method
simsdOptionsOption set for simsd
predictPredict K-step ahead model output
predictOptionsOption set for predict
forecastForecast identified model output
forecastOptionsOption set for forecast
idinputGenerate input signals

Response Computation and Visualization

simSimulate response of identified model
bodeBode plot of frequency response, or magnitude and phase data
bodeplotPlot Bode frequency response with additional plot customization options
bodemagBode magnitude response of LTI models
stepStep response plot of dynamic system; step response data
stepplotPlot step response and return plot handle
stepinfoRise time, settling time, and other step-response characteristics
nyquistNyquist plot of frequency response
nyquistplotNyquist plot with additional plot customization options
impulseImpulse response plot of dynamic system; impulse response data
impulseplotPlot impulse response and return plot handle
pzmapPole-zero plot of dynamic system
pzplotPole-zero map of dynamic system model with plot customization options
iopzmapPlot pole-zero map for I/O pairs of model
iopzplotPlot pole-zero map for I/O pairs and return plot handle
spectrumOutput power spectrum of time series models
spectrumplotPlot disturbance spectrum of linear identified models
lsimSimulate time response of dynamic system to arbitrary inputs
lsimplotSimulate response of dynamic system to arbitrary inputs and return plot handle
lsiminfoCompute linear response characteristics
showConfidenceDisplay confidence regions on response plots for identified models
findstatesEstimate initial states of model
data2stateMap past data to states of state-space and nonlinear ARX models
simOptionsOption set for sim
stepDataOptionsOptions set for step
bodeoptionsCreate list of Bode plot options
nyquistoptionsList of Nyquist plot options
timeoptionsCreate list of time plot options
getoptionsReturn @PlotOptions handle or plot options property
setoptionsSet plot options for response plot
pzoptionsCreate list of pole/zero plot options
spectrumoptionsOption set for spectrumplot
identprefSet System Identification Toolbox preferences
findstatesOptionsOption set for findstates

Time Series Analysis

arEstimate parameters of AR model for scalar time series
armaxEstimate parameters of ARMAX model using time-domain data
arxEstimate parameters of ARX or AR model using least squares
etfeEstimate empirical transfer functions and periodograms
spaEstimate frequency response with fixed frequency resolution using spectral analysis
spafdrEstimate frequency response and spectrum using spectral analysis with frequency-dependent resolution
ivarAR model estimation using instrumental variable method
n4sidEstimate state-space model using subspace method
ssestEstimate state-space model using time or frequency domain data
pemPrediction error estimate for linear and nonlinear model
nlarxEstimate parameters of nonlinear ARX model
idpolyPolynomial model with identifiable parameters
idssState-space model with identifiable parameters
idnlarxNonlinear ARX model
getpvecModel parameters and associated uncertainty data
setpvecModify value of model parameters
initSet or randomize initial parameter values
noise2measNoise component of model
spectrumOutput power spectrum of time series models
forecastForecast identified model output
simSimulate response of identified model
arOptionsOption set for ar
forecastOptionsOption set for forecast
simOptionsOption set for sim

Online Estimation

Online Parameter Estimation

recursiveARCreate System object for online parameter estimation of AR model
recursiveARMACreate System object for online parameter estimation of ARMA model
recursiveARXCreate System object for online parameter estimation of ARX model
recursiveARMAXCreate System object for online parameter estimation of ARMAX model
recursiveBJCreate System object for online parameter estimation of Box-Jenkins polynomial model
recursiveOECreate System object for online parameter estimation of Output-Error polynomial model
recursiveLSCreate System object for online parameter estimation using recursive least squares algorithm
stepUpdate model parameters and output online using recursive estimation algorithm
cloneCopy online parameter estimation System object
resetReset online parameter estimation System object
releaseUnlock online parameter estimation System object
isLockedLocked status of online parameter estimation System object
rpemEstimate general input-output models using recursive prediction-error minimization method
rplrEstimate general input-output models using recursive pseudolinear regression method
segmentSegment data and estimate models for each segment

Online State Estimation

extendedKalmanFilterCreate extended Kalman filter object for online state estimation
unscentedKalmanFilterCreate unscented Kalman filter object for online state estimation
particleFilterParticle filter object for online state estimation
correctCorrect state and state estimation error covariance using extended or unscented Kalman filter, or particle filter and measurements
predictPredict state and state estimation error covariance at next time step using extended or unscented Kalman filter, or particle filter
initializeInitialize the state of the particle filter
cloneCopy online state estimation object
Was this topic helpful?