Conceptos básicos de la identificación de modelos lineales
Los modelos lineales son los modelos más sencillos que se pueden identificar utilizando System Identification Toolbox™. Utilice la identificación de modelos lineales cuando un modelo lineal es suficiente para capturar por completo la dinámica del sistema. Para identificar modelos lineales, debe comenzar con datos de entrada-salida del dominio del tiempo o del dominio de la frecuencia y una estructura de modelo, como un modelo de espacio de estados o de función de transferencia. El software ajusta iterativamente los parámetros libres del modelo para minimizar la diferencia entre la salida medida y la respuesta del modelo simulada a los datos de entrada. La toolbox permite realizar las siguientes tareas:
Realizar la estimación de modelos lineales utilizando una estructura de modelo específica.
Utilizar un enfoque de modelado de caja negra y explorar qué estructura de modelo es más adecuada para los datos.
Construir un modelo lineal preliminar y utilizarlo para inicializar los parámetros del modelo que desea estimar.
Incorporar conocimiento del sistema al modelo fijando los parámetros conocidos en valores específicos.
Utilizar estimación regularizada para reducir la incertidumbre del modelo restringiendo la flexibilidad del modelo.
Temas
Identificar modelos lineales
- Identify Linear Models Using System Identification App
Identify linear black-box models from single-input/single-output (SISO) data using the System Identification app. - Identify Linear Models Using the Command Line
Identify linear models from multiple-input/single-output (MISO) data using System Identification Toolbox commands. - Frequency Domain Identification: Estimating Models Using Frequency Domain Data
This example shows how to estimate models using frequency domain data. - Estimation Report
The estimation report contains information about the results and options used for a model estimation.
Seleccionar una estructura de modelo
- About Identified Linear Models
System Identification Toolbox software uses objects to represent a variety of linear and nonlinear model structures. - Available Linear Models
Summary of linear model types that you can use for system identification. - Black-Box Modeling
Black-box modeling is useful when your primary interest is in fitting the data regardless of a particular mathematical structure of the model. - Model Structure Selection: Determining Model Order and Input Delay
This example shows some methods for choosing and configuring the model structure. - Modeling Multiple-Output Systems
Use a multiple-output modeling technique that suits the complexity and internal input-output coupling of your system. - Types of Model Objects
Model object types include numeric models, for representing systems with fixed coefficients, and generalized models for systems with tunable or uncertain coefficients.
Estructuras y restricciones de objetos de modelo
- Linear Model Structures
Linear models in System Identification Toolbox take the form of model objects that are linear model structures. You can construct model objects directly or use estimation commands to both construct and estimate models. You can also modify the properties of existing model objects. - Imposing Constraints on Model Parameter Values
Constrain the adjustments that the estimation algorithm can make to individual model parameters by using theStructureproperty of the mode object.
Regularización
- Regularized Identification of Dynamic Systems
This example shows the benefits of regularization for identification of linear and nonlinear models. - Estimate Regularized ARX Model Using System Identification App
This example shows how to estimate regularized ARX models using automatically generated regularization constants in the System Identification app. - Regularized Estimates of Model Parameters
Regularization is the technique for specifying constraints on the flexibility of a model, thereby reducing uncertainty in the estimated parameter values.
Temas adicionales
- Loss Function and Model Quality Metrics
Configure the loss function that is minimized during parameter estimation. After estimation, use model quality metrics to assess the quality of identified models. - Effect of Input Intersample Behavior on Continuous-Time Models
The intersample behavior of the input signals influences the estimation, simulation and prediction of continuous-time models.
Ejemplos destacados
Data and Model Objects in System Identification Toolbox
Manage data and model objects available in the System Identification Toolbox.
Estimating Simple Models from Real Laboratory Process Data
Develop and analyze simple models from a real laboratory process data.
