Standard

Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases. / Wang, Rongxiao; Chen, Bin; Qiu, Sihang; Zhu, Zhengqiu; Wang, Yiduo; Wang, Yiping; Qiu, Xiaogang.

In: International Journal of Environmental Research and Public Health, Vol. 15, No. 7, 1450, 2018, p. 1-19.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Wang, R, Chen, B, Qiu, S, Zhu, Z, Wang, Y, Wang, Y & Qiu, X 2018, 'Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases' International Journal of Environmental Research and Public Health, vol. 15, no. 7, 1450, pp. 1-19. https://doi.org/10.3390/ijerph15071450

APA

Wang, R., Chen, B., Qiu, S., Zhu, Z., Wang, Y., Wang, Y., & Qiu, X. (2018). Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases. International Journal of Environmental Research and Public Health, 15(7), 1-19. [1450]. https://doi.org/10.3390/ijerph15071450

Vancouver

Wang R, Chen B, Qiu S, Zhu Z, Wang Y, Wang Y et al. Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases. International Journal of Environmental Research and Public Health. 2018;15(7):1-19. 1450. https://doi.org/10.3390/ijerph15071450

Author

Wang, Rongxiao ; Chen, Bin ; Qiu, Sihang ; Zhu, Zhengqiu ; Wang, Yiduo ; Wang, Yiping ; Qiu, Xiaogang. / Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases. In: International Journal of Environmental Research and Public Health. 2018 ; Vol. 15, No. 7. pp. 1-19.

BibTeX

@article{e681175a09f14ad3a599fd2cd876a9d7,
title = "Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases",
abstract = "Dispersion prediction plays a significant role in the management and emergency response to hazardous gas emissions and accidental leaks. Compared with conventional atmospheric dispersion models, machine leaning (ML) models have both high accuracy and efficiency in terms of prediction, especially in field cases. However, selection of model type and the inputs of the ML model are still essential problems. To address this issue, two ML models (i.e., the back propagation (BP) network and support vector regression (SVR) with different input selections (i.e., original monitoring parameters and integrated Gaussian parameters) are proposed in this paper. To compare the performances of presented ML models in field cases, these models are evaluated using the Prairie Grass and Indianapolis field data sets. The influence of the training set scale on the performances of ML models is analyzed as well. Results demonstrate that the integrated Gaussian parameters indeed improve the prediction accuracy in the Prairie Grass case. However, they do not make much difference in the Indianapolis case due to their inadaptability to the complex terrain conditions. In addition, it can be summarized that the SVR shows better generalization ability with relatively small training sets, but tends to under-fit the training data. In contrast, the BP network has a stronger fitting ability, but sometimes suffers from an over-fitting problem. As a result, the model and input selection presented in this paper will be of great help to environmental and public health protection in real applications.",
keywords = "Back propagation network, Field case, Hazardous gas dispersion prediction, Input selection, Support vector regression",
author = "Rongxiao Wang and Bin Chen and Sihang Qiu and Zhengqiu Zhu and Yiduo Wang and Yiping Wang and Xiaogang Qiu",
year = "2018",
doi = "10.3390/ijerph15071450",
language = "English",
volume = "15",
pages = "1--19",
journal = "International Journal of Environmental Research and Public Health",
issn = "1660-4601",
publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",
number = "7",

}

RIS

TY - JOUR

T1 - Comparison of Machine Learning Models for Hazardous Gas Dispersion Prediction in Field Cases

AU - Wang, Rongxiao

AU - Chen, Bin

AU - Qiu, Sihang

AU - Zhu, Zhengqiu

AU - Wang, Yiduo

AU - Wang, Yiping

AU - Qiu, Xiaogang

PY - 2018

Y1 - 2018

N2 - Dispersion prediction plays a significant role in the management and emergency response to hazardous gas emissions and accidental leaks. Compared with conventional atmospheric dispersion models, machine leaning (ML) models have both high accuracy and efficiency in terms of prediction, especially in field cases. However, selection of model type and the inputs of the ML model are still essential problems. To address this issue, two ML models (i.e., the back propagation (BP) network and support vector regression (SVR) with different input selections (i.e., original monitoring parameters and integrated Gaussian parameters) are proposed in this paper. To compare the performances of presented ML models in field cases, these models are evaluated using the Prairie Grass and Indianapolis field data sets. The influence of the training set scale on the performances of ML models is analyzed as well. Results demonstrate that the integrated Gaussian parameters indeed improve the prediction accuracy in the Prairie Grass case. However, they do not make much difference in the Indianapolis case due to their inadaptability to the complex terrain conditions. In addition, it can be summarized that the SVR shows better generalization ability with relatively small training sets, but tends to under-fit the training data. In contrast, the BP network has a stronger fitting ability, but sometimes suffers from an over-fitting problem. As a result, the model and input selection presented in this paper will be of great help to environmental and public health protection in real applications.

AB - Dispersion prediction plays a significant role in the management and emergency response to hazardous gas emissions and accidental leaks. Compared with conventional atmospheric dispersion models, machine leaning (ML) models have both high accuracy and efficiency in terms of prediction, especially in field cases. However, selection of model type and the inputs of the ML model are still essential problems. To address this issue, two ML models (i.e., the back propagation (BP) network and support vector regression (SVR) with different input selections (i.e., original monitoring parameters and integrated Gaussian parameters) are proposed in this paper. To compare the performances of presented ML models in field cases, these models are evaluated using the Prairie Grass and Indianapolis field data sets. The influence of the training set scale on the performances of ML models is analyzed as well. Results demonstrate that the integrated Gaussian parameters indeed improve the prediction accuracy in the Prairie Grass case. However, they do not make much difference in the Indianapolis case due to their inadaptability to the complex terrain conditions. In addition, it can be summarized that the SVR shows better generalization ability with relatively small training sets, but tends to under-fit the training data. In contrast, the BP network has a stronger fitting ability, but sometimes suffers from an over-fitting problem. As a result, the model and input selection presented in this paper will be of great help to environmental and public health protection in real applications.

KW - Back propagation network

KW - Field case

KW - Hazardous gas dispersion prediction

KW - Input selection

KW - Support vector regression

UR - http://www.scopus.com/inward/record.url?scp=85050031679&partnerID=8YFLogxK

UR - http://resolver.tudelft.nl/uuid:e681175a-09f1-4ad3-a599-fd2cd876a9d7

U2 - 10.3390/ijerph15071450

DO - 10.3390/ijerph15071450

M3 - Article

VL - 15

SP - 1

EP - 19

JO - International Journal of Environmental Research and Public Health

T2 - International Journal of Environmental Research and Public Health

JF - International Journal of Environmental Research and Public Health

SN - 1660-4601

IS - 7

M1 - 1450

ER -

ID: 46760807