Use this link to cite:
http://hdl.handle.net/2183/31478 Modelado experimental de dos sistemas de laboratorio multivariables
Loading...
Identifiers
Publication date
Authors
Rivero-Contreras, Rogelio
Zamarreño, Jesús
Tadeo, Fernando
Advisors
Other responsabilities
Journal Title
Bibliographic citation
Rivero-Contreras R., Zamarreño J.M., Tadeo F. Modelado experimental de dos sistemas de laboratorio multivariables. XLIII Jornadas de Automática: libro de actas, pp.599-606. https://doi.org/10.17979/spudc.9788497498418.0599
Type of academic work
Academic degree
Abstract
[Resumen] Se presenta la obtención de modelos en función de transferencia discreta de dos sistemas de laboratorio multivariables. La identificación del sistema se realiza en base a datos experimentales obtenidos mediante la realización de pruebas tipo escalón secuenciales, simultáneas y rotadas. Se utiliza la toolbox HIDEN, disponible para MATLAB, considerando los métodos del error de la salida y mínimos cuadrados. Una vez obtenidos los modelos, se realiza la validación mediante los criterios establecidos en los índices de Akaike (AIC), error de predicción final (FPE) y longitud mínima de Rissanen (RMDL). La validación de los modelos indica que los de mayor precisión se obtienen a partir de entradas secuenciales.
[Abstract] The identification of discrete transfer function models of two multivariable laboratory systems is presented. This identification is conducted based on experimental data obtained by performing sequential, simultaneous, and rotated step input tests. The HIDEN toolbox is used, considering the output error and least squares methods. Once the models are obtained, validation is performed comparing the models obtained using the Akaike (AIC), final prediction error (FPE) and minimum Rissanen length (RMDL) indexes. This comparison indicates that the most accurate models are obtained from sequential inputs in both cases.
[Abstract] The identification of discrete transfer function models of two multivariable laboratory systems is presented. This identification is conducted based on experimental data obtained by performing sequential, simultaneous, and rotated step input tests. The HIDEN toolbox is used, considering the output error and least squares methods. Once the models are obtained, validation is performed comparing the models obtained using the Akaike (AIC), final prediction error (FPE) and minimum Rissanen length (RMDL) indexes. This comparison indicates that the most accurate models are obtained from sequential inputs in both cases.
Description
Editor version
Rights
Atribución-NoComercial-CompartirIgual 4.0 Internacional (CC BY-NC-SA 4.0)
https://creativecommons.org/licenses/by-nc-sa/4.0/deed.es


