Real-time Discrete Nonlinear Identification via Recurrent High Order Neural Networks Identificación No Lineal en Tiempo Real usando Redes Neuronales
نویسندگان
چکیده
This paper deals with the discrete-time nonlinear system identification via Recurrent High Order Neural Networks, trained with an extended Kalman filter (EKF) based algorithm. The paper also includes the respective stability analysis on the basis of the Lyapunov approach for the whole scheme. Applicability of the scheme is illustrated via real-time implementation for a three phase induction motor.
منابع مشابه
Real-time Discrete Nonlinear Identification via Recurrent High Order Neural Networks Identificación No Lineal en Tiempo Real usando Redes Neuronales Recurrentes de Alto Orden
This paper deals with the discrete-time nonlinear system identification via Recurrent High Order Neural Networks, trained with an extended Kalman filter (EKF) based algorithm. The paper also includes the respective stability analysis on the basis of the Lyapunov approach for the whole scheme. Applicability of the scheme is illustrated via real-time implementation for a three phase induction motor.
متن کاملUna Comparación entre Estrategias Evolutivas y RPROP para la Estimación de Redes Neuronales
Resumen—Rprop ha sido reconocido como uno de los más poderosos algoritmos para entrenar redes neuronales artificiales; sin embargo, el algoritmo de estrategias de evolución es un fuerte competido para resolver problemas de optimización debido a su capacidad para buscar el óptimo global sin la necesidad de usar información sobre el gradiente. En este artículo, se comparan ambos algoritmos usando...
متن کاملModelado de Cambios de Nivel en Series de Tiempo No Lineales Usando Redes Neuronales Artificiales
It is usually accepted that many physical and economical variables shown nonlinear dynamical behaviors which complexity makes it impossible to formulate an econometric model based only on physical or economic laws that properly represents its evolution. The formulation of model is aggravated with the presence of outliers and structural breaks, for which there are not any mathematical models tha...
متن کاملUn Modelo para la Prediccion de Recidiva de Pacientes Operados de Cancer de Mama (CMO) Basado en Redes Neuronales
La predicción de recidiva en pacientes que han sido operados de cáncer de mama juega un papel muy importante en tareas médicas como el diagnostico y la planificación del tratamiento que hay que realizarle al mismo. En la actualidad, los expertos médicos están llevando a cabo estas tareas usando técnicas no numéricas. Las redes neuronales artificiales se muestran como una herramienta potente par...
متن کاملReal-time Discrete Nonlinear Identification via Recurrent High Order Neural Networks
This paper deals with the discrete-time nonlinear system identification via Recurrent High Order Neural Networks, trained with an extended Kalman filter (EKF) based algorithm. The paper also includes the respective stability analysis on the basis of the Lyapunov approach for the whole scheme. Applicability of the scheme is illustrated via real-time implementation for a three phase induction motor.
متن کامل