Bayesian neural networks on the inference of distillation product quality

The control of the distillation process in oil refineries requires the evaluation of product quality throughout the operation of the plant. This paper uses Bayesian neural networks, combined to several pre-processing and variable selection techniques, to develop systems for inferencing the quality o...

Full description

Saved in:
Bibliographic Details
Main Authors: Barbosa, C.H., Melo, B., Vellasco, M., Pacheco, M., Vasconcellos, L.P.
Format: Conference Proceeding
Language:English
Subjects:
Backpropagation
Bayesian methods
Control theory
Input variables
Neural networks
Neurons
Pattern recognition
Principal component analysis
Rivers
Testing
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The control of the distillation process in oil refineries requires the evaluation of product quality throughout the operation of the plant. This paper uses Bayesian neural networks, combined to several pre-processing and variable selection techniques, to develop systems for inferencing the quality of distillation products, for REPAR refinery (Refinaria do Parana), operated by PETROBRAS.
DOI:10.1109/SBRN.2002.1181447