Proposta de um algoritmo GPC adaptativo com baixo custo computacional
Nenhuma Miniatura disponível
Data
2015-02-10
Autores
Mazoco, Bruna Marques
Título da Revista
ISSN da Revista
Título de Volume
Editor
Universidade Federal do Espírito Santo
Resumo
This dissertation proposes a new formulation of the Generalized Predictive Control (GPC) algorithm to be applied in industrial plants with time-varying models. The unitary control horizon premisse alows it to be simple enough to be implemented in a Programmable Logic Controller (PLC). The explicit solution of the control increment is obtained from the parameters of the model and the GPC tuning parameters (prediction horizon hp and supression weight ), in addition to past and present inputs and outputs. Supression weight tuning is done by Root Locus technique, constructed from the system closed loop characteristic polynomium, everytime the model parameters of the industrial plant (stable or not in open loop) su er modi fication.
Descrição
Palavras-chave
Sintonia de parâmetros