Modèles de corrélation
L'analyse de corrélation dans le domaine temporel fait référence à l'estimation non paramétrique de la réponse impulsionnelle des systèmes dynamiques en tant que modèle de réponse impulsionnelle finie (FIR) à partir des données. Le modèle estimé est stocké sous la forme d’un objet de fonction de transfert.
Applications
System Identification | Identify models of dynamic systems from measured data |
Fonctions
cra | Estimate impulse response using input/output data prewhitening before correlation analysis |
impulseest | Nonparametric impulse response estimation |
impulseestOptions | Options set for impulseest |
era | Estimate state-space model from impulse response data using Eigensystem Realization Algorithm (ERA) (depuis R2022b) |
getpvec | Obtain model parameters and associated uncertainty data |
setpvec | Modify values of model parameters |
Rubriques
- What Is Time-Domain Correlation Analysis?
Time-domain correlation analysis refers to nonparametric estimation of the impulse response of dynamic systems as a finite impulse response (FIR) model from the data.
- Estimate Impulse-Response Models Using System Identification App
Estimate in the app using time-domain correlation analysis.
- Estimate Impulse-Response Models at the Command Line
Use
impulseest
command to estimate using correlation analysis. - Compute Response Values
Obtain numerical impulse- and step-response vectors as a function of time.
- Identify Delay Using Transient-Response Plots
You can use transient-response plots to estimate the input delay, or dead time, of linear systems.