Controle de sistemas não lineares atraves de redes neurais
AUTOR(ES)
Edilberto Pereira Teixeira
DATA DE PUBLICAÇÃO
1991
RESUMO
This thesis presents an approach for controlling a nonlinear system using an adaptive scheme implemented through neural networks. For a class of systems. necessary conditions for teaching the system input-output mapping to a multi-layer neural network trained by the generalized deIta rule are presented. A feedback linearization procedure, using feedforward neural networks, is also introduced in this work. The method can be applied for unknown continuous time systems as well as discrete time or sampled systems. The feedback linearization procedure can be either adaptively accomplished, having the neural network weights adjusted during normal operation of the plant, or in a non adaptive mode, having the weights adjusted off-line. In the proposed scheme, two neural networks are trained in two different stages. In the first stage, the nonlinear system is exclted several times to teach the inverse dynamics of the system to a neural network. In the second stage. the system is again excited several times to train a second neural network with input signals that desired way
ASSUNTO(S)
industrias - automação engenharia de sistemas
ACESSO AO ARTIGO
http://libdigi.unicamp.br/document/?code=vtls000033829Documentos Relacionados
- Controle adaptativo de sistemas dinâmicos não-lineares utilizando realimentação de saída e redes neurais artificiais.
- Identificação e controle de processos não lineares utilizando redes neurais artificiais
- Uso de redes neurais e conhecimento a priori na identificação de sistemas dinâmicos não lineares.
- Identificação de sistemas não-lineares multivariáveis usando redes neurais perceptron multicamadas e função de base radial
- Otimização de sistemas atraves de redes neurais artificiais