2008
DOI: 10.1016/j.conengprac.2007.01.003
|View full text |Cite
|
Sign up to set email alerts
|

Data reconciliation: A robust approach using a contaminated distribution

Abstract: On-line optimisation provides a means for maintaining a process around its optimum operating range. This optimisation heavily relies on process measurements and accurate process models. However, these measurements often contain random and possibly gross errors as a result of miscalibration or failure of the measuring instruments. This paper proposes a data reconciliation strategy that deals with the presence of such gross errors. Instead of constructing the objective function to be minimized on the basis of ra… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
24
0
4

Year Published

2008
2008
2019
2019

Publication Types

Select...
4
2
1

Relationship

1
6

Authors

Journals

citations
Cited by 48 publications
(29 citation statements)
references
References 17 publications
0
24
0
4
Order By: Relevance
“…It is interesting to note that the knowledge of parameter distribution errors is not a limiting factor, indeed it would be easy to modify the algorithm in order to run without any parameter knowledge. The next step would be to include some robustness properties with regard gross measurement errors using a so-called contaminated distribution in the maximum likelihood estimation [2], [1]. Finally the proposed method must be evaluated on real process model with real data in terms of process model adjustment along the successive heat to observe the impact of the setpoints adjustment on the successive batch processes.…”
Section: Discussionmentioning
confidence: 99%
“…It is interesting to note that the knowledge of parameter distribution errors is not a limiting factor, indeed it would be easy to modify the algorithm in order to run without any parameter knowledge. The next step would be to include some robustness properties with regard gross measurement errors using a so-called contaminated distribution in the maximum likelihood estimation [2], [1]. Finally the proposed method must be evaluated on real process model with real data in terms of process model adjustment along the successive heat to observe the impact of the setpoints adjustment on the successive batch processes.…”
Section: Discussionmentioning
confidence: 99%
“…Este problema foi proposto originalmente por Alhaj-Dibo et al (2008), sendo constituído por 16 correntes, cada uma das quais caracterizada pela vazão e por duas composições (que não necessariamente somam 1, pois deve existir mais componentes). Exceto pelas vazões das correntes: 1, 4 e 11, todas as demais variáveis são medidas.…”
Section: O Problema De Processamento De Minériounclassified
“…Comparou-se o desempenho do estimador Normal Contaminada utilizado por Alhaj-Dibo et al (2008) e o estimador Lorenziana, presente no pacote de reconciliação do simulador EMSO. O resultados mostraram melhor desempenho do estimador Lorenziana, o que corroborou o estudo prévio de Prata et al (2008) que ressaltou o desempenho deste estimador e o de Welsch.…”
Section: Conclusõesunclassified
See 2 more Smart Citations