Application of Generative Adversarial Network for the Prediction of Gasoline Properties

Application of Generative Adversarial Network for the Prediction of Gasoline Properties

Near-infrared (NIR) spectroscopy has been widely used to predict the gasoline properties that are difficult to measure online during gasoline blending. NIR models should be prepared in advance to apply this technique successfully. Obtaining a high-accuracy NIR model in practice is hard because abund...

Full description

Bibliographic Details
Main Authors: Kaixun He, Jingjing Liu, Zhi Li
Format: Article
Language:English
Published: AIDIC Servizi S.r.l. 2020-08-01
Series:Chemical Engineering Transactions
Online Access:https://www.cetjournal.it/index.php/cet/article/view/11093
id doaj-3faa720e0f984d0f8d7085bbcc201c8d
record_format Article
spelling doaj-3faa720e0f984d0f8d7085bbcc201c8d2021-02-16T11:27:38ZengAIDIC Servizi S.r.l.Chemical Engineering Transactions2283-92162020-08-018110.3303/CET2081152Application of Generative Adversarial Network for the Prediction of Gasoline PropertiesKaixun HeJingjing LiuZhi LiNear-infrared (NIR) spectroscopy has been widely used to predict the gasoline properties that are difficult to measure online during gasoline blending. NIR models should be prepared in advance to apply this technique successfully. Obtaining a high-accuracy NIR model in practice is hard because abundant labelled samples are difficult to acquire. A new modelling method on the basis of Wasserstein generative adversarial network is proposed in this study to overcome this weakness. Abundant artificial labelled samples are generated firstly using the proposed method, and sample selection is performed to select the appropriate artificial samples. Real and selected artificial samples from the selection results are combined to train the NIR model that could be established efficiently when labelled samples are scarce. An actual dataset obtained during gasoline blending is provided to validate the effectiveness of the proposed method, and several traditional methods are adopted for comparison.https://www.cetjournal.it/index.php/cet/article/view/11093
collection DOAJ
language English
format Article
sources DOAJ
author Kaixun He
Jingjing Liu
Zhi Li
spellingShingle Kaixun He
Jingjing Liu
Zhi Li
Application of Generative Adversarial Network for the Prediction of Gasoline Properties
Chemical Engineering Transactions
author_facet Kaixun He
Jingjing Liu
Zhi Li
author_sort Kaixun He
title Application of Generative Adversarial Network for the Prediction of Gasoline Properties
title_short Application of Generative Adversarial Network for the Prediction of Gasoline Properties
title_full Application of Generative Adversarial Network for the Prediction of Gasoline Properties
title_fullStr Application of Generative Adversarial Network for the Prediction of Gasoline Properties
title_full_unstemmed Application of Generative Adversarial Network for the Prediction of Gasoline Properties
title_sort application of generative adversarial network for the prediction of gasoline properties
publisher AIDIC Servizi S.r.l.
series Chemical Engineering Transactions
issn 2283-9216
publishDate 2020-08-01
description Near-infrared (NIR) spectroscopy has been widely used to predict the gasoline properties that are difficult to measure online during gasoline blending. NIR models should be prepared in advance to apply this technique successfully. Obtaining a high-accuracy NIR model in practice is hard because abundant labelled samples are difficult to acquire. A new modelling method on the basis of Wasserstein generative adversarial network is proposed in this study to overcome this weakness. Abundant artificial labelled samples are generated firstly using the proposed method, and sample selection is performed to select the appropriate artificial samples. Real and selected artificial samples from the selection results are combined to train the NIR model that could be established efficiently when labelled samples are scarce. An actual dataset obtained during gasoline blending is provided to validate the effectiveness of the proposed method, and several traditional methods are adopted for comparison.
url https://www.cetjournal.it/index.php/cet/article/view/11093
work_keys_str_mv AT kaixunhe applicationofgenerativeadversarialnetworkforthepredictionofgasolineproperties
AT jingjingliu applicationofgenerativeadversarialnetworkforthepredictionofgasolineproperties
AT zhili applicationofgenerativeadversarialnetworkforthepredictionofgasolineproperties
_version_ 1724267770955169792

Similar Items