Algoritmo para linealización de controladores fuzzy
; Franco Domínguez, Samuel Otros responsables
Fecha
03/2017
Editorial
Universidad Católica de Santiago del Estero. Departamento Académico San Salvador
Revista
Difusiones
Título del libro
Nombre del evento
Fecha del evento
Institución organizadora
ISSN
2314-1662
eISSN
ISBN
Idioma
Español
Tipo del evento
Clasificación temática
Palabras clave
Resumen
Los controladores basados en lógica fuzzy utilizan heurísticas de control derivadas del conocimiento de un experto que no requieren estrictamente del conocimiento profundo del proceso al que se aplican. El diseño específico de controladores fuzzy sigue un procedimiento general que deja algunas consideraciones bajo el criterio del diseñador. En este trabajo se establece como criterio principal, que un controlador fuzzy genere una curva de transferencia prácticamente lineal, para no interferir con el proceso de control aplicado a plantas generalmente no lineales. Se analizan diferentes tipos de funciones de pertenencia, alcances de las variables fuzzy y el efecto que producen las distribuciones no uniformes en las particiones fuzzy para obtener una curva de transferencia prácticamente lineal, considerando que el proceso de inferencia fuzzy es básicamente alineal. El algoritmo propuesto se aplica para diseñar controladores fuzzy, que se incorporan en un modelo de control de un mezclador de caudales en línea, produciendo excelentes resultados.
Fuzzy logic based controllers use control heuristics derived from knowledge of an expert who does not strictly require deep knowledge of the process to which it is applied. The specific design of fuzzy logic controllers follows a general procedure that leaves some considerations at the discretion of the designer. This work is set as the main criterion, a fuzzy controller generates a practically linear transfer curve, to not interfere with the control process applied to usually non-linear plants. Different types of membership functions, scope of fuzzy variables and the effect produced by non-uniform distributions in fuzzy partitions are analyzed, to obtain a substantially linear transfer curve, considering the fuzzy inference process is basically nonlinear. The proposed algorithm is applied to design fuzzy logic controllers, which are incorporated in a control model of a flow mixer in line, producing excellent results.
Fuzzy logic based controllers use control heuristics derived from knowledge of an expert who does not strictly require deep knowledge of the process to which it is applied. The specific design of fuzzy logic controllers follows a general procedure that leaves some considerations at the discretion of the designer. This work is set as the main criterion, a fuzzy controller generates a practically linear transfer curve, to not interfere with the control process applied to usually non-linear plants. Different types of membership functions, scope of fuzzy variables and the effect produced by non-uniform distributions in fuzzy partitions are analyzed, to obtain a substantially linear transfer curve, considering the fuzzy inference process is basically nonlinear. The proposed algorithm is applied to design fuzzy logic controllers, which are incorporated in a control model of a flow mixer in line, producing excellent results.
Aprobación
Revisión
Complementado por
Referenciado por
Licencia Creative Commons
Excepto donde se indique lo contrario, la licencia de este ítem se describe como by-nc-sa
