System Identification Toolbox

Preface

What Is the System Identification Toolbox?

Using This Guide

Typographical Conventions

Related Products

About the Author

The System Identification Problem

Basic Questions About System Identification

Common Terms Used in System Identification

Basic Information About Dynamic Models

The Signals

The Basic Dynamic Model

Variants of Model Descriptions

How to Interpret the Noise Source

Terms to Characterize the Model Properties

The Basic Steps of System Identification

A Startup Identification Procedure

Step 1: Looking at the Data

Step 2: Getting a Feel for the Difficulties

Step 3: Examining the Difficulties

Step 4: Fine Tuning Orders and Disturbance Structures

Multivariable Systems

Reading More About System Identification

The Graphical User Interface

The Big Picture

The Model and Data Boards

The Working Data

The Views

The Validation Data

The Work Flow

Management Aspects

Workspace Variables

Help Texts

Handling Data

Data Representation

Getting Input-Output Data into the GUI

Taking a Look at the Data

Preprocessing Data

Checklist for Data Handling

Simulating Data

Estimating Models

The Basics

Direct Estimation of the Impulse Response

Direct Estimation of the Frequency Response

Estimation of Parametric Models

ARX Models

ARMAX, Output-Error and Box-Jenkins Models

State-Space Models

User Defined Model Structures

Examining Models

Views and Models

The Plot Windows

Frequency Response and Disturbance Spectra

Transient Response

Poles and Zeros

Compare Measured and Model Output

Residual Analysis

Text Information

LTI Viewer

Further Analysis in the MATLAB Workspace

Some Further GUI Topics

Mouse Buttons and Hotkeys

Troubleshooting in Plots

Layout Questions and idprefs.mat

Customized Plots

What Cannot be Done Using the GUI

Tutorial

Overview

The Toolbox Commands

An Introductory Example to Command Mode

The System Identification Problem

Impulse Responses, Frequency Functions, and Spectra

Polynomial Representation of Transfer Functions

State-Space Representation of Transfer Functions

Continuous-Time State-Space Models

Estimating Impulse Responses

Estimating Spectra and Frequency Functions

Estimating Parametric Models

Subspace Methods for Estimating State-Space Models

Data Representation and Nonparametric Model Estimation

Data Representation

Correlation Analysis

Spectral Analysis

More on the Data Representation in iddata

Parametric Model Estimation

ARX Models

AR Models

General Polynomial Black-Box Models

State-Space Models

Optional Variables

Defining Model Structures

Polynomial Black-Box Models: The idpoly Model

Multivariable ARX Models: The idarx Model

Black-Box State-Space Models: the idss Model

Structured State-Space Models with Free Parameters: the idss Model

State-Space Models with Coupled Parameters: the idgrey Model

State-Space Structures: Initial Values and Numerical Derivatives

Examining Models

Parametric Models: idmodel and its children

Frequency Function Format: the idfrd model

Graphs of Model Properties

Transformations to Other Model Representations

Discrete and Continuous Time Models

Model Structure Selection and Validation

Comparing Different Structures

Impulse Response to Determine Delays

Checking Pole-Zero Cancellations

Residual Analysis

Model Error Models

Noise-Free Simulations

Assessing the Model Uncertainty

Comparing Different Models

Selecting Model Structures for Multivariable Systems

Dealing with Data

Offset Levels

Outliers and Bad Data; Multi-Experiment Data

Missing Data

Filtering Data: Focus

Feedback in Data

Delays

Recursive Parameter Estimation

The Basic Algorithm

Choosing an Adaptation Mechanism and Gain

Available Algorithms

Segmentation of Data

Some Special Topics

Time Series Modeling

Periodic Inputs

Connections Between the Control System Toolbox and the System Identification Toolbox

Memory/Speed Trade-Offs

Local Minima

Initial Parameter Values

Initial State

The Estimated Parameter Covariance Matrix

No Covariance

nk and InputDelay

Linear Regression Models

Spectrum Normalization and the Sampling Interval

Interpretation of the Loss Function

Enumeration of Estimated Parameters

Complex-Valued Data

Strange Results

Function Reference

Functions -- By Category

Help Functions

The Graphical User Interface

Simulation and Prediction

Data Manipulation

Nonparametric Estimation

Parameter Estimation

Model Structure Creation

Manipulating Model Structures

Model Conversions

Model Analysis

Model Validation

Assessing Model Uncertainty

Model Structure Selection

Recursive Parameter Estimation

General

Functions -- Alphabetical List

Preface