In the paper, a method for improving the performance of data-driven Soft Sensors in the presence of small datasets problems is proposed. The method is based on a preliminary feature extraction phase, followed by a data selection method. Selected data are used to train neural-network-based Soft Sensors, modeling the featureoutput nonlinear relationship. A case of study referring to a Sour Water Stripping plant is used in the paper to validate the suitability of the proposed methodology.
Titolo: | Improving soft sensors performance in the presence of small datasets by data selection |
Autori: | |
Data di pubblicazione: | 2020 |
Abstract: | In the paper, a method for improving the performance of data-driven Soft Sensors in the presence of small datasets problems is proposed. The method is based on a preliminary feature extraction phase, followed by a data selection method. Selected data are used to train neural-network-based Soft Sensors, modeling the featureoutput nonlinear relationship. A case of study referring to a Sour Water Stripping plant is used in the paper to validate the suitability of the proposed methodology. |
Handle: | http://hdl.handle.net/11570/3180529 |
ISBN: | 978-1-7281-4460-3 |
Appare nelle tipologie: | 14.d.3 Contributi in extenso in Atti di convegno |
File in questo prodotto:
Non ci sono file associati a questo prodotto.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.