Modelagem computacional de dados e controle inteligente no espaço de estado / State space computational data modelling and intelligent control

AUTOR(ES)
DATA DE PUBLICAÇÃO

2005

RESUMO

This study presents contributions for state space multivariable computational data modelling with discrete time invariant as well as with time varying linear systems. A proposal for Deterministic-Estocastica Modelling of noisy data, MOESP_AOKI Algorithm, is made. We present proposals forsolving the Discrete-Time Algebraic Riccati Equation as well as the associate Linear Matrix Inequalityusing a multilayer Recurrent Neural Network approaches. An Intelligent Linear Parameter Varying(ILPV) control approach for multivariable discrete Linear Time Varying (LTV) systems identified bythe MOESP_VAR algorithm, are both proposed. A gain scheduling adaptive control scheme based on neural networks is designed to tune on-line the optimal controllers. In synthesis, an Intelligent Linear Parameter Varying (ILPV) Control approach for multivariable discrete Linear Time Varying Systems (LTV), identified by the algorithm MOESP_VAR, is proposed. This way an Intelligent LPV Control for multivariable data computationally modeled via the MOESP_VAR algorithm is structured, implemented and tested with good results.

ASSUNTO(S)

modelagem de dados data modeling sistemas lineares - identificação espaço e tempo subspace methods sistemas mimo sistemas de tempo discreto redes neurais (computação) series temporais discrete-time systems time varing linear systems intelligent control sistemas inteligentes de controle gainscheduling . mimo systems neural networks state space time series

Documentos Relacionados