User`s guide

2 Choosing Your System Identification Strategy
Recommended Model Estimation Sequence
System identificatio n is an iterat iv e process, wh ere you identify models
with different structures from data and compare model performance. You
start by estimating the parameters of simple model structures. If the model
performance is poor, you g radually increase the complexity of the model
structure. Ultimately, you choose the simplest model that best describes
the d ynam ics of your system .
Another reason to start with simple model structures is that higher-order
models are not always more accurate. Increasing model complexity increases
the uncertainties in parameter e stimates and typically requires more data
(which is co mmon in the case of nonlinear models).
Note Model structure is not the only factor that determines model accuracy.
If your model is p oor, you might need to preprocess your data by removing
outliers or filtering n ois e. For more information, see “Ways to Process Data
for System Identification” on page 1-2.
Estimate impulse-response and frequency-response models first to gain
insight into the system dynamics and assess whether a linear model is
sufficient. Then, estimate parametric models in the following order:
1 ARX polynomial and state-space models provide the simplest structures.
These m odels let you estimate the model orde r and noise dynamics.
In the System Identification Tool GUI. Select to es timate the ARX
linear parametric model and the state-space model using the N4SID
method.
At the comm
and line.Usethe
arx and the n4sid commands.
For more information, see “Identify ing Input-Output Polynomial Models”
on page 3-41 and “Identifying State-Space Models” on page 3-73.
2 ARMAX and BJ polynomial models provide more complex structures and
require iterative estimation. Try several model orders and keep the model
orders as low as possible.
2-2