Mathworks SYSTEM IDENTIFICATION TOOLBOX RELEASE NOTES Release Notes

System Identification Toolbox™ Release Notes

How to Contact MathWorks

www.mathworks. comp.soft-sys.matlab Newsgroup www.mathworks.com/contact_TS.html Technical Support

suggest@mathworks.com Product enhancement suggestions

bugs@mathwo doc@mathworks.com Documentation error reports service@mathworks.com Order status, license renewals, passcodes

info@mathwo

com

rks.com

Web

Bug reports

Sales, prici

ng, and general information

508-647-7000 (Phone)

508-647-7001 (Fax)

The MathWorks, Inc. 3 Apple Hill Drive Natick, MA 01760-2098

For contact information about worldwide offices, see the MathWorks Web site.

System Identification Toolbox™ Release Notes

The software desc ribed in this document is furnished under a license agreeme nt. The software may be used or copied only under the terms of the license agreement. No part of this manual may be photocopied or reproduced in any form without prior written consent from The MathWorks, Inc.

FEDERAL ACQUISITION: This provision applies to all acquisitions of the Program and Documentation by, for, or through the federal government of the United States. By accepting delivery of the Program or Docum entation, the government hereby agrees that this software or documentation qualifies as commercial computer software or commercial computer software documentation as such terms are used or defined in FAR 12.212, DFARS Part 227.72, and DFARS 252.227-7014. Accordingly, the terms and conditions of this Agreement and only those rights specified in this Agreement, shall pertain to and govern theuse,modification,reproduction,release,performance,display,anddisclosureoftheProgramand Documentation by the federal government (or other entity acquiring for or through the federal governme nt) and shall supersede any conflicti ng contractual terms or conditions. If this License fails to meet the government’s needs or is inconsistent in any respect with federal procurement law, the government agrees to return the Program and Documentation, unused , to The MathWorks, Inc.

Trademarks

MATLAB and Simulink are registered trademarks of The M athWorks, Inc. See

www.mathworks.com/trademarks for a list of additional trademarks. Other product or brand

names may be trademarks or registered trademarks of their respective holders.

Patents

MathWorks products are protected by one or more U.S. patents. Please see

www.mathworks.com/patents for more information.

Summary by Version ............................... 1

Version 7.4.1 (R2010b) System Identification Toolbox

Software

Version 7.4 (R2010a) System Identification Toolbox

Software

Version 7.3.1 (R2009b) System Identification Toolbox

Software

Version 7.3 (R2009a) System Identification Toolbox

Software

Version 7.2.1 (R2008b) System Identification Toolbox

Software

........................................ 5

........................................ 6

........................................ 9

........................................ 10

........................................ 12

Contents

Version 7.2 (R2008a) System Identification Toolbox

Software

Version 7.1 (R2007b) System Identification Toolbox

Software

Version 7.0 (R2007a) System Identification Toolbox

Software

Version 6.2 (R2006b) System Identification Toolbox

Software

Version 6.1.3 (R2006a) System Identification Toolbox

Software

Version 6.1.2 (R14SP3) System Identification Toolbox

Software

........................................ 13

........................................ 20

........................................ 21

........................................ 24

........................................ 26

........................................ 28

iii

Version 6.1.1 (R14SP2) System Identification Toolbox

Software

Version 6.0 (R13SP2) System Identification Toolbox

Software

Compatibility S umm ary for System Identification

Toolbox Software

........................................ 29

........................................ 30

................................ 34

iv Contents

SummarybyVersion

This table provides quick access to w h at’s new in each version. For clarification, see “Using Release No tes” on page 2.

System Identification Toolbox™ Release Notes

Version (Release)

Latest Versi V7.4.1 (R201

V7.4 (R2010a)

V7.3.1 (R2009b)

V7.3 (R2009a)

V7.2.1

V7.2 (R2008a)

V7.1 (R2007b)

V7.0

(R2008b)

(R2007a)

on 0b)

New Features and Changes

No No Bug Reports

Yes Details

No No Bug Repor

Yes Details

No Yes

Yes Details

Yes

Detai

Yes Details

Version Compatibilit Consideratio

Yes Summary

No Bug Repo

Summary

Yes Summary

No Bug Re

No Bug Reports No

Fixed Bugs

and Known

Problems

Includes fix

Bug Reports No

ports

rts

Related Documentation a t Web Site

Printable Release Notes: PDF

Current product documentation

System Identification Toolbox™ Release Notes

Version 7.4 (R2010a) System Identification Toolbox Software

This table summarizes what’s new in Version 7.4 (R2010a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features introduced in this version:

• “New Ability to Use Discrete-Time Linear Models for Nonlinear Black-Box

Estimation” on page 6

• “New Cell Array Support for B and F Polynomials of Multi-Input

Polynomial Models” on page 7

• “Functions and Function Elements Being Removed” on page 8

Fixed Bugs and Known Problems

Bug Reports No

Related Documentation a t Web Site

New Ability to Use Discrete-Time Linear Models for Nonlinear Black-Box Estimation

You can now use the following discrete-time linear models for initi al izing a nonlinear black-box estimation.

Discrete-time Linear Model

Single-output polynomial model of ARX structure (

idpoly)

Use for Initializing...

Single-output nonlinear ARX model estimation

Multi-output polynomial model of ARX structure (

idarx)

Multi-output nonlinear ARX model estimation

Version 7.4 (R2010a) System Iden tification Toolbox™ So ftware

Discrete-time Linear Model

Single-output polynomial model of Output-Error (OE) structure (

idpoly) or state-space model with

no disturbance component ( object with K = 0

State-space model with no disturbance component ( object with K = 0)

During estimation, the software uses the linear model orders and delay as initial values of the nonlinear model orders and delay. For nonlinear ARX models, this initialization always provides a better fit to the estimation data than the linear ARX model.

You can use a linear model as an alternative approach to using model orders and delay for nonlinear estimation of the same system.

You can estimate or construct the linear model and then use this model for constructing (see

nlhw) the nonlinear model. For more information, see “Using Linear

Model for Nonlinear ARX Estimation”, and “Using Linear M odel for Hammerstein-Wiener Estimation” in the System Identification Toolbox User’s Guide.

idnlarx and idnlhw) or estimating (see nlarx or

idss)

idss

Use for Initializing...

Single-output Hammerstein-Wiener model estimation

Multi-output Hammerstein-Wiener model estimation

New Cell Array Support for B and F Polynomials of Multi-Input Polynomial M odels

You can now use cell arrays to specify the B and F polynomials of multi-input polynomial models. The B and F polynomials are represented by the properties of an idpoly object These properties are currently double matrices.

For multi-input polynomial models, these polynomials will be represented by cell arrays only in a future version. If your code performs operations on the and f properties, make one of the following changes in the code:

b and f

System Identification Toolbox™ Release Notes

• When you construct the model using the idpoly command, use cell arrays

to specify the B and F polynomials. Using cell arrays causes the properties to be represented by cell array s.

b and f

• After you construct or estimate the model, use the new

When you use cell arrays, you must also update your code to use cell array syntax on

Note For single-input polynomial models, the b and f properties continue to be double row vectors.

Functions and Function Elements Being Removed

Function or Function Element Name

Double matrix support for

b and f properties

of multi-input models.

idpoly

setPolyFormat

command to:

- Convert b and f properties to cell arrays.

- Make the model backward compatible to continue using double matrices

for

b and f properties. This operation ensures that operations on b and

f properties that use matrix syntax continue to work without errors

in a future version.

b and f properties instead of matrix syntax.

What Happens When you Use the Function or Element?

Warns

UseThisInstead Compatibility

Considerations

Use cell array to specify the and f properties of multi-in put polynomial models.

If your code performs operations on the

b and f properties,

update the code to be compatible with a future release. See “New Cell Array Support for B and F Polynomials of Multi-Input Polynomial Models” on page 7.

Version 7.3.1 (R2009b) System Identification Toolbox™ Software

Version 7.3.1 (R2009b) System Identification Toolbox Software

This table summarizes what’s new in Version 7.3.1 (R2009b):

New Features and Changes

No No Bug Reports No

Version Compatibility Considerations

Fixed Bugs and Known Problems

Related Documentation a t Web Site

System Identification Toolbox™ Release Notes

Version 7.3 (R2009a) System Identification Toolbox Software

This table summarizes what’s new in Version 7.3 (R2009a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports

New features introduced in this version:

• “Enhanced Handling of Offsets and Trends in Signals” on page 10

• “Ability to Get Regressor Values in Nonlinear ARX Models” on page 11

Fixed Bugs and Known Problems

Related Documentation a t Web Site

Printable Release Notes: PDF

Current product documentation

Enhanced Handling of Offsets and Trends in Signals

This version of the product includes new and expanded functionality for handling offsets and trends in signals. This data processing operation is necessary for estimating more accurate linear models because linear models cannot capture arbitrary differences between the input and output signal levels.

The previous vers ion of the product let you remove mean values or linear trends from steady-state signals using the GUI and the For transient signals, you had to remove offsets and trends using matrix manipulation.

detrend function.

The GUI functionality for removing means and linear trends from signals is unchanged. However, you can now do the following at the command line:

• Save the values of means or linear trends removed during detrending using

anew

detrend output argument. You can use this saved trend information

to detrend other data sets. You can also restore subtracted trends to the

Version 7.3 (R2009a) System Iden tification Toolbox™ So ftware

output simulated by a linear model that was estimated from detrended data.

For example, this syntax computes and removes mean values from the data, and saves these values to the output variable

[data_d,T]=detrend(data). T is an object with properties that store offset

and slope information for input and output signals.

• Remove any offset or linear trend from the data using a new

detrend

input argument. This is useful for removing arbitrary nonzero offsets from transient data or applying previously saved trend information to any data set.

For example, this syntax removes an offset or trend specified by

= detrend(data,T)

T: data_d

• Add an arbitrary offset or linear trend to data signals. This is useful when

you want to simulate the response of a linear model about a nonzero equilibrium input-output level and this model was estimated from detrended data.

For example, this syntax adds trend information to a simulated model output

y_sim,whichisaniddata object: y = retrend(y_sim,T). T

specifies the offset an d slope information for in puts and outputs.

For more information, see “Handling Offsets and Trends in Data”.

Ability to Get Regressor Values in Nonlinear ARX Models

The getreg command can now return the numerical values of regressors in nonlinear ARX models and provides an intermediate output of nonlinear ARX m odels .

This advanced functionality converts input and output values to regressors, and passes the regressor values to the

evaluate command to compute

the model response. This incremental step lets you gain insight into the propagation of information through the nonlinear ARX model.

For more information, see the

getreg reference page. To learn more about

the nonline ar ARX model structure, see “Nonlinear B lack-Box Model Identification”.

System Identification Toolbox™ Release Notes

Version 7.2.1 (R2008b) System Identification Toolbox Software

This table summarizes what’s new in Version 7.2.1 (R2008b):

New Features and Changes

No Yes

Version Compatibility Considerations

Summary

Functions and Properties Being Rem oved

For more information about the process of removing functions, see "About Functions and Properties Being Remove d" in “What Is in the Release Notes” on page 2.

Function or Property Name

model.Algorithm. Trace

What Happens When You Us e Function or Property?

Still runs

Fixed Bugs and Known Problems

Bug Reports No

Use This Instead Compatibility

model.Algorithm. Display

Related Documentation a t Web Site

Considerations

Using

model.Algorithm. Trace

results in a warning.

Version 7.2 (R2008a) System Iden tification Toolbox™ So ftware

Version 7.2 (R2008a) System Identification Toolbox Software

This table summarizes what’s new in Version 7.2 (R2008a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features introduced in this version are:

• “Simulating Non linear Black-Box Models in Simulink Software” on page 13

• “Linearizing Nonlinear Black-Box Models a t User-Specified Operating

Points” on page 14

• “Estimating Multiple-Output Models Using Weighted Sum of Least

Squares Minimization Criterion” o n page 15

• “Improved H andling of Initial States for Linear and Nonlinear Models”

on page 16

• “Improved Algorithm Options for L i n ear Models” on page 17

• “New Block Reference Pages” on page 18

• “Functions and Properties Being Removed” on page 18

Fixed Bugs and Known Problems

Bug Reports No

Related Documentation a t Web Site

Simulating Nonlinear Black-Box Models in Simulink Software

You can now simulate nonlinear ARX and Hammerstein-Wiener models in Simulink using the nonlinear ARX and the Hammerstein-Wiener model blocks in the System Identification Toolbox™ block library. This is useful in the following situations:

• Representing dynamics of a physical component in a Simulink model using

a data-based nonlinear model

System Identification Toolbox™ Release Notes

• Replacing a complex Simulink sub system with a simpler data- ba s ed

nonlinear model

Note Nonlinear ARX Model and Hammerstein-Wiener Model blocks read variables from the MATLAB (base) workspace or model workspace. When the MATLAB workspace and model workspace contain a variable with the same name and this variable is referenced by a Simulink block, the variable in the model workspace takes precedence.

IfyouhaveinstalledReal-TimeWorkshop®software, you can generate code from models containing nonlinear ARX and the Hammerstein-Wiener model blocks. However, you cannot generate code when:

• Hammerstein-Wiener models use the

output nonlinearity.

• Nonlinear ARX models use custom regressors or use the

neuralnet nonlinearity estimator.

You can access the new System Identification Toolbox blocks from the Simulink Library Browser. For more information about these blocks, see the IDNLARX Model (nonlinear ARX model) and the IDNLHW Model (Hammerstein-Wiener model) block reference pages.

customnet estimator for input or

customnet or

Linearizing Nonlinear Black-Box Models at User-Specified Operating Points

You can now use the linearize command to linearize nonlinear black-box models, including nonlinear ARX and Hammerstein-W iener models, at specified operating poin ts. Linearization produces a first-order Taylor series approximation of the system about an operating point. An operating point is defined by the set of constant input and state values for the model.

Ifyoudonotknowtheoperatingpoint,youcanusethe compute it from specifications, such as steady-state requirements or values of these quantities at a given time instant from the simulation of the model.

findop command to

Version 7.2 (R2008a) System Iden tification Toolbox™ So ftware

For nonlinear ARX models, if all of the steady-state input and output values are known, you can map these values to the model state values using the

data2state command.

linearize replaces lintan and removes the restriction for linearizing models

containing custom regressors or specific nonlinearity estimators, such as

neuralnet and treepartition.

If you have installed Simulink

Control Design™ software, you can linearize nonlinear ARX and Hammerstein-Wiener models in Simulink after importing them into Simulink.

For more information, see:

• “Linear Approximation of Nonlinear Black-Box Models” about comput i ng

operating points and linearizing models

• “Simulating Model Output” about importing nonlinear black-bo x models

into Simulink

Estimating Multiple-Output Models Using Weighted Sum of Least Squares Minimization Criterion

You can now specify a custom weighted trace criterion for minimization when estimating linear and nonlinear black-box models for multiple-output systems. This feature is useful for controlling the relative importance of output channels during the estimation process.

The

Algorithm property of linear and nonlinear models now provides the

Criterion field for choosing the minimization criterion. This new field can

have the following values:

•

det — (Defau lt) Specify this option to minimize th e determinant of the

prediction error covariance. This choice leads to maximum likelihood estimates of model parameters. It implicitly uses the inverse of estimated noise variance as the weighting function. This option was already available in previous releases.

•

trace — Specify this option to define your own weighing function that

controls the relative weights of output signals during the estimation. This criterion minimizes the weighted sum of least square prediction errors. You

System Identification Toolbox™ Release Notes

can specify the relative weighting of prediction errors for each output using the new

Weighting field of the Algorithm property. B y default, Weighting

is an identity matrix, which means that all outputs are weighed e qually. Set

Weighting to a positive semidefinite symmetric m atrix.

For more information about these new Algorithm fields for linear estimation, see the

Algorithm Properties reference page. For more information about

Algorithm fields for nonlinear estimation, see the reference pages.

Note If you are estimating a single-output model, det and tr ace values of the

Criterion field produce the same estimation results.

Improved Handling of Initial States for Linear and Nonlinear Models

The following are new options to handle initial states for nonlinear models:

idnlarx and idnlhw

• For n onlinear ARX models (

vector for initial states when using

idnlarx), you can now specify a numerical

sim or predi ct by setting the Init

argument. For example:

predict(model,data,'init',[1;2;3;4])

where the last argument is the state vector.

For more information, see the

• For Hammerstein-Wiener models (

the initial states when using

For more information, see the

sim and predict reference pages.

idnlhw), you can now choose to estimate

predict or nlhw by setting INIT='e'.

predict and nlhw reference pages.

If you want to specify your own initial states, s ee the corresponding model reference pages for a definition of the states for each model type.

If you do not know the states, you can use the

findop or the findstates

command to compute the states. For more information about using these commands, see the

findstates(idnlarx),andfindstates(idnlhw) reference pages.

findop(idnlarx), findop(idnlhw),

Version 7.2 (R2008a) System Iden tification Toolbox™ So ftware

To help you interpret the states of a nonlinear ARX model, you can use the

getDelayInfo command. For more information, see the get DelayInfo

reference page.

The

findstates command is available for all linear and nonlinear models.

Also see the

findstates(idmodel) and findstates(idnlgrey) reference

pages.

Improved Algorithm Options for Linear Models

The following improvements are available for the Algorithm property of linear models to align linear and nonlinear models (where appropriate) and improve robustness for default settings:

• The

SearchDirection field (model.Algorithm.SearchDirection)has

been renamed to

SeachMethod (model.Algorithm.SearchMethod)tobe

consistent with the nonlinear models, where the corresponding field is

SeachMethod.

• The new

model.Algorithm.SearchMethod='lsqnonlin' uses the lsqnonlin

lsqnonlin option for specify ing SearchMethod is available.

optimizer from the Optimization Toolbox™ software. You must have Optimization Toolbox software installed to use this option.

• The improved

for specifying

gn algorithm (subspace Gauss-Newton method) is available

SearchDirection. The updated gn algorithm better handles

the scale of the parameter Jacobians and is also consistent with the algorithm used for nonlinear model estimation.

• The default values for th e

(

modelname.Algorithm.LimitError) is changed to 0, which is consistent

LimitError field of the Algorithm property

with the corresponding option f or estimating no n linear models. In previous releases,

LimitError default value was 1.6, which robustified

the estimation process against data outliers by associating a linear penalty for large errors, rather than a quadratic penalty. Now, there is no robustification by default ( the default setting and plot the prediction errors using

LimitError=0). You can estimate the model with

pe(data.model).If

the resulting plot s hows occasional large values, repeat the estimation with

model.Algorithm.LimitError set to a value between 1 and 2.

• The

model.Algorithm.Advanced property has a new tolerance field

GnPinvConst corresponding to the gn SearchMethod. GnPinvConst

System Identification Toolbox™ Release Notes

specifies that singular values of the Jacobian that are sm aller tha n

GnPinvConst*max(size(J))*norm(J)*eps are discarded when computing

the search direction. You can assign a p ositive real value for this field. Default value is

1e4.

• The d efault value of

has been changed from possibility of a situation where the estimation algorithm does not converge (predictor becomes unstable) while still allowing enough flexibility to capture lightly damped modes.

For more information about Algorithm properties of linear models, see the

Algorithm Properties reference page.

New Block Reference Pages

New documentation for System Identification Toolbox blocks is provided. For more information, see “Block Reference” in the System Identification Toolbox reference documentation.

Functions and Properties Being Rem oved

For more information about the process of removing functions, see "About Functions and Properties Being Remove d" in “What Is in the Release Notes” on page 2.

Function or Property Name

lintan

What Happens When You Us e Function or Property?

Still runs

model.Algorithm.Advanced.Zstability property

1.01 to 1+sqrt(eps). The new default reduces the

Use This Instead Compatibility

Considerations

linearize(idnlhw) linearize(idnlarx)

See “Linearizing Nonlinear Black-Box Models at User-Specifi ed Operating Points” on page 14.

Version 7.2 (R2008a) System Iden tification Toolbox™ So ftware

Function or Property Name

model.Algorithm. SearchDirection

gns option of model.Algorithm. SearchDirection

GnsPinvTol of model.Algorithm.Advanced

What Happens When You Us e Function or Property?

Still runs

Use This Instead Compatibility

Considerations

model.Algorithm. SearchMethod

See “Improved Algorithm Options for Linear Models” on page 17.

GnPinvConst

See “Improved Algorithm Options for Linear Models” on page 17.

System Identification Toolbox™ Release Notes

Version 7.1 (R2007b) System Identification Toolbox Software

This table summarizes what’s new in Version 7.1 (R2007b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports No

New feature introduced in this version:

Fixed Bugs and Known Problems

Related Documentation a t Web Site

New Polynomial Nonlinearity Estimator for Hammerstein-Wiener Models

You can now estimate nonlinearities for Hammerstein-Wiener models using a single-variable polynom ial at either the input or the output. This nonlinearity estimator is available at the command line.

For more information, see the information about estimating Hammerstein-Wiener models, see “Identifying Hammerstein-Wiener Models ” in the System Identifica t ion Toolbox documentation.

poly1d reference pages. For more

Version 7.0 (R2007a) System Iden tification Toolbox™ So ftware

Version 7.0 (R2007a) System Identification Toolbox Software

This table summarizes what’s new in Version 7.0 (R2007a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports No

New features and changes introduced in this version are:

• “New N onlinear Black-Box Modeling Options” on page 21

• “New Nonlinear Grey-Box Modeling Option” on page 22

• “New Getting Started Guide” on page 23

• “Revised and Expanded User’s Guide” on page 23

Fixed Bugs and Known Problems

Related Documentation a t Web Site

New Nonlinear Black-Box Modeling Options

You can now estimate nonlinear discrete-time black-box models for both single-output an d multiple-out put time-do main data. The System Identification Toolbox product supports the following types of nonlinear black-box models:

• Hammerstein-Wiener

• Nonlinear ARX

To learn how to estimate nonlinear b lack-box models using the System Identification Tool GUI or commands in the MATLAB Command Window, see the System Identification Toolbox documentation.

Note You can estimate Hammerstein-Wiener black-box models from input-output data only. These models do not support time-series data, where thereisnoinput.

System Identification Toolbox™ Release Notes

New demos are available to help you explore nonlinear black-box functions. For more information, see the collection of demos in the Tutorials on Nonlinear ARX and Hammerstein-Wiener Model Identification category.

New Nonlinear Grey-Box Modeling Option

You can now estimate nonlinear discrete-time and continuous-time models for arbitrary nonlinear ordinary differential equations using single-output and multiple-output time-domain data, or time-series data (no measured inputs). Models that you can specify as a set of nonlinear ordinary differential equations (ODEs) are called grey-box models.

To learn how to estimate nonlinear grey-box models using the co mmands in the MATLAB Co mmand Window, see System Identification Toolbox documentation.

Specify the OD E in a function or a MEX-file. The template file for writing the MEX-file,

matlab/toolbox/ident/nlident.

IDNLGREY_MODEL_TEMPLATE.c, is located in

To estimate the equation parameters, first construct an specify the ODE file and the parameters you want to estimate. Use estimate the ODE parameters. For more information, see the

pem reference pages.

New demos are available to help you e xplore nonlinear grey-box functions. For more information, see the collection ofdemosintheTutorialsonNonlinear Grey-Box Model Identification category.

idnlgrey object to

pem to

idnlgrey and

Optimization Toolbox Search Method for Nonlinear Estimation Is Supported

If you have Optimization Toolbox software installed, you can specify the

lsqnonlin search method for estimating black-box and grey-box nonlinear

models in the MATLAB Command Window.

model.algorithm.searchmethod='lsqnonlin'

For more information, see the idnlarx, idnlhw,andidnlgrey reference pages.

Version 7.0 (R2007a) System Iden tification Toolbox™ So ftware

New Getting Star

The System Ident Started Guide. T provides the fo

• “Tutorial – Id

the System Ide linear black-

• “Tutorial – I

Using the GUI graphical u fit single-

• “Tutorial –

Tutorial f and method

Revised a

The Syste expanded

m Identification Toolbox docum entation has been revised and

ification Toolbox product now provides a new Getting

his guide introduces fundamental identification concepts and

llowing tutorials to help you get started quickly:

entifying Linear Models Using the GUI” — Tutorial for using

ntification Tool grap hi ca l user interface (GUI) to estimate

box models for single-input and single-output (SISO) data.

dentifying Low-Order Transfer Functions (Process Models)

” — Tutorial for using the System Identification Tool

ser interface (GUI) to estimate low-o rde r transfer functions to

input and single-output (SISO) d ata.

Identifying Linear Models Using the Command Line” —

or estimating models using System Identification Toolbox objects

s for multiple-input and single-output (MISO) data.

nd Expanded User’s Guide

ted Guide

System Identification Toolbox™ Release Notes

Version 6.2 (R2006b) System Identification Toolbox Software

This table summarizes what’s new in Version 6.2 (R2006b):

New Features and Changes

Yes Details below

Version Compatibility Considerations

No Bug Reports No

New feature introduced in this version:

Fixed Bugs and Known Problems

Related Documentation a t Web Site

MATLAB Compiler Support

The System Identification Too lbox product now supports the MATLAB Compiler™ product.

YoucanuseMATLABCompilertotakeMATLABfilesasinputandgenerate redistributable, standalone applications that include System Identification Toolbox functionality, including the following:

• Creating data and model objects

• Preprocessing and manipulating data

• Simulating models

• Transforming models, including conversions between continuous and

discrete time and model reduction

• Plotting transient and frequency response

To use these features, write a function that uses System Identification Toolbox commands. Use the MATLAB Compiler software to create a standalone application from the MATLAB Compiler file. F or more information, see the MATLAB Compiler documentation.

Version 6.2 (R2006b) System Iden tification Toolbox™ So ftware

Standalone applications that include System Identification Toolbox functionality have the following limitations:

• No access to the System Identification library in the Simulink software

(

slident)

• No support for model estimation

System Identification Toolbox™ Release Notes

Version 6.1.3 (R2006a) System Identification Toolbox Software

This table summarizes what’s new in Version 6.1.3 (R2006a):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Details below. See also Summary.

New features and changes introduced in this version are:

• “balred Introduced for Model Reduction” on page 26

• “Search Direction for Minimizing Criteria Can Be Computed by Adaptive

Gauss-Newton Method” on page 26

• “Maximum Number of Bisections Used by Line Search Is Increased” o n

page 27

Fixed Bugs and Known Problems

Bug Reports No

Related Documentation a t Web Site

balred Introduced for Model Reduction

Use balred to perform model re duction instead of idmodred.

Search Direction for Minimizing Criteria Can Be Computed by Adaptive Gauss-Newton Method

An adaptive Gauss-Newton method is now available for computing the direction of the line search for cost-function minimization. Use this method when you observe convergence problems in the estimation results, or as an alternative to the Levenberg-Marquard (

lm)method.

The

gna search method was suggested by Adrian Wills, Brett Ninness, and

Stuart Gibson in their paper "On Gradient-Based Search for Multivariable System Estimates", presented at the IFAC World Congress in Prague in 2005.

gna is an adaptive version of gns and uses a cutoff value for the singular

values of the criterion Hessian, which is adjusted adaptively depending on thesuccessofthelinesearch.

Version 6.1.3 (R2006a) System Identification Toolbox™ Software

Specify the gna method by setting the Search Direction property to 'gna'. For example:

m = pem(data,model_structure,'se','gna')

The default initial value of gamma in the gna search is 10^-4.Youcanseta different value using the

SearchDirection,seetheAlgorithm Properties reference pages.

InitGnaTol property. For more information about

Maximum Number of Bisections Used by Line Search Is Increased

The default value for the MaxBisections property, which is the maximum number of bisections along the search direction used by line search, is increased from criterion value along the search vector.

10 to 25. This increases the number of attempts to find a lower

For more information about

Properties

Functions and Properties Being Rem oved

For more information about the process of removing functions, see "About Functions and Properties Being Remove d" in “What Is in the Release Notes” on page 2.

Function or Property Name

idmodred

reference page.

What Happens When You Us e Function or Property?

Still runs

Search properties, see the Algorithm

Use This Instead Compatibility

Considerations

balred

See “balred Introduced for Model Reduction” on page 26.

System Identification Toolbox™ Release Notes

Version 6.1.2 (R14SP3) System Identification Toolbox Software

This table summarizes what’s new in Version 6.1.2 (R14SP3):

New Features and Changes

No No Bug Reports No

Version Compatibility Considerations

Fixed Bugs and Known Problems

Related Documentation a t Web Site

Version 6.1.1 (R14SP2) System Identification Toolbox™ Software

Version 6.1.1 (R14SP2) System Identification Toolbox Software

This table summarizes what’s new in Version 6.1.1 (R14SP2):

New Features and Changes

No No

Version Compatibility Considerations

Fixed Bugs and Known Problems

Fixed bugs

Related Documentation a t Web Site

System Identification Toolbox™ Release Notes

Version 6.0 (R13SP2) System Identification Toolbox Software

This table summarizes what’s new in Version 6.0 (R13SP2):

New Features and Changes

Yes Details below

Version Compatibility Considerations

Yes Summary

New features and changes introduced in this version are:

• “idproc Model Object Added” on page 30

• “Estimation and Vali dation in Frequency Domain Now Supported” on

page 31

• “Continuous-Time Data Can Now Be Stored Using Frequency-Domain

Objects” on page 31

• “Simulink Software Now Supports iddata and idmodel Objects” on page 32

• “advice About Data and Models Now Available” on page 32

• “Theta Models No Longer Supported” on page 32

Fixed Bugs and Known Problems

No bug fixes

Related Documentation a t Web Site

V6.0 product documentation

idproc Model Object Added

A new model object, idproc, is used to represent simple continuous-time process models. This object is characterized by static gain, possible dead time, and dominating time constant(s). A new GUI that supports this object is available in the System Identification Toolbox GUI.

To learn more about this object, type run a demo.

You can also try the comman d

m = pem(data,'p1d')

iddemopr at the MATLAB prompt to

Version 6.0 (R13SP2) System Identification Toolbox™ Software

Estimation and V Supported

You can now perfo such as the foll

• Inputs and out

• Frequency-re

Both System I (GUI) suppor

All estimat data and fre frequency

Use the domains. U frequenc

Type at th

help iddata

ion, simulation, and validation routines accept frequency-domain

-response data must be packaged as an

fft

y-dependent resolution.

e MATLAB prompt:

rm estimation and validation using frequency-domain data,

owing:

puts, entered as frequency-domain data in the

sponse data from a frequency analyzer

dentification Toolbox functions and the graphical user interface

tthis.

quency-response data as inputs, similar to time-domain data. The

and ifftfunctions to transform between the time and frequency

se the

spafdr function to estimate frequency responses using

alidation in Frequency Domain Now

iddata object

frd or idfrd object.

idprops data

mplete descriptions. To access a demo, type

for co

Cont Freq

You c Cont arb For

inuous-Time Data Can Now Be Stored Using uency-Domain Objects

an now store continuous-time data as a frequency-domain data object.

inuous-time Fourier-transformed data is now stored at a finite number of

itrary frequencies, letting you estimate continuous-time models directly.

example, type at the MATLAB prompt:

help oe

iddemofr.

System Identification Toolbox™ Release Notes

Simulink Software Now Supports iddata and idmodel Objects

You can now import and simulate System Identification Toolbox idmodel models in the Simulink environment. You can also import iddata objects into Simulink.

The command can use to simulate contains data sou rces and sinks for

slident opens a System Identification block library, which you

idmodel models (with or without noise). This library also

iddata objects.

advice About Data and Models Now Available

Use the new advice command to get h elpful tips about the quality, problems, and options associated with an

For more information, type at the MATLAB prompt:

help iddata/advice

and

help idmodel/advice

iddata or idmodel obje ct.

Theta Models No Longer Supported

Theta models (matrices) are no longer supported.

Functions and Properties Being Rem oved

For more information about the process of removing functions, see "About Functions and Properties Being Remove d" in “What Is in the Release Notes” on page 2.

Version 6.0 (R13SP2) System Identification Toolbox™ Software

Function or Property Name

th2par

th2ss

What

Use This Instead Compatibility Happens When You Us e Function or Property?

Errors None

Still runs

None

Considerations

See “Theta Models No Longer Supported” on page 32.

System Identification Toolbox™ Release Notes

Compatibility Summary for System Identification Toolbox Software

This table summarizes new features and changes that might cause incompatibilities when you upgrade from an earlier version, or when you use files on multiple versions. Details are provided with the description of the new feature or change.

Version (Release) New Features and Changes with Version

Compatibility Impact

Latest Version V7.4.1 (R2010b)

V7.4 (R20010a) See “Functions and Function Elements Being

V7.3.1 (R2009b)

V7.3 (R2009a)

V7.2.1 (R2008b) See “Functions and Properties Being

V7.2 (R2008a) See “Functions and Properties Being

V7.1 (R2007b)

V7.0 (R2007a)

V6.2 (R2006b)

V6.1.3 (R2006a) See “Functions and Properties Being

V6.1.2 (R14SP3)

V6.1.1 (R14SP2)

V6.0 (R13SP2) See “Functions and Properties Being

None

Removed” on page 8.

None

Removed” on page 12.

Removed” on page 18.

None

Removed” on page 27.

None

Removed” on page 32.

Mathworks SYSTEM IDENTIFICATION TOOLBOX RELEASE NOTES Release Notes

Specifications and Main Features

Frequently Asked Questions

User Manual