Estimação parametrica robusta atraves de redes neurais artificiais

AUTOR(ES)
DATA DE PUBLICAÇÃO

1995

RESUMO

Artificial Neural Networks can achieve high computation rates by employing a massive number of simple processing elements with a high degree of connectivity between these elements. Neural networks with feedback connections provide a computing model to solve a rich class of optimization problems. This dissertation presents an application of Hopfield s Neural Networks in Robust Parametric Estimation with unknown-but-bounded disturbance. The Discret Hopfield s Network is used to calculate a parameter uncertainty set for model parameters. Any element in this set can be considered a good estimator for the real parameters. A Modified Hopfield s Network has also been described and it is useful for getting efficient and reliable sets. Comparative analysis with others robust estimation approaches are included. The Valid-Subspace technique is used to obtain the internal parameters of the Hopfield s Neural Network. These parameters are explicitlycomputed, based upon problem specifications, to assure the network convergence. In this case, the equilibrium point represents a solution to robust estimation problem with unknown-but-bounded error

ASSUNTO(S)

inteligencia artificial redes neurais (computação)

Documentos Relacionados