Please use this identifier to cite or link to this item:
|Type:||Artigo de evento|
|Title:||Predictive Hybrid Model For Pulping Process And Pulp Properties|
|Author:||Aguiar Helena Cristina|
Filho Rubens Maciel
Pasquali Silvia Maria
De Jesus Reinaldo
|Abstract:||A hybrid model is proposed as an alternative to first principles and pure neural network modeling techniques with the intent of improving accuracy and extrapolation ability. Preliminary results show that a simple hybrid network, which uses the deterministic model result to reinforce the neural net mapping, is able to improve model prediction as well as training time, even when a limited training set is available.|
|Editor:||Appita Inc, Carlton, Australia|
|Appears in Collections:||Unicamp - Artigos e Outros Documentos|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.