Título: Identification and Inverse Modelling of Industrial Systems Using Regional Models
Autores: Antonio Galeno Pereira Neto, Luis Gustavo Mota Souza and Francisco de Assis da Silva Mota.
Resumo:
With the advancement of technology the speed of industrial processes has greatly increased resulting in the need of obtaining models and controllers in a faster and more interactive way. Fortunately, the speed and ability to obtain data have also shown great advances, allowing the use of techniques capable of modeling processes reliably and quickly using the System Identification process. For generate a model from the input and output data of the systems, the System Identification has been the subject of many studies, with several techniques being proposed capable of generating reliable models in a short period of time. Two of these techniques, presented in this article, are the techniques known as Regional Models and Robust Regional Models, which use Clustering techniques such as Self Organizing Map (SOM) and K-means to dividing the system’s data space into similar regions in order to produce more reliable models using supervised neural networks; the robust approach also performs the treatment of Outliers in the data using the M-Estimation technique. The techniques presented are applied in nonlinear industrial systems and evaluated based on their Normalized Mean Square Error (NMSE) and the residual autocorrelation.
Palavras-chave:
System Identification, Neural Networks, ELM, SOM, Regional Models.
Páginas: 8
Código DOI: 10.21528/CBIC2021-63
Artigo em pdf: CBIC_2021_paper_63.pdf
Arquivo BibTeX: CBIC_2021_63.bib