Bayesian network-based model for assessing the intensity of outbursts of coal and gas
by Wen Nie; Hongwei Yang; Hailong Zhang; Jiabo Geng; Xin Wu
International Journal of Oil, Gas and Coal Technology (IJOGCT), Vol. 13, No. 2, 2016

Abstract: Fully understanding the major parameters contributing to an outburst is significant in the evaluation of coal and gas outbursts. However, these parameters can also affect each other which significantly increase the difficulty of evaluation. Considering related uncertainties, a Bayesian network (BN) which reflects major variable parameters affects the object without knowing the clear relations between the parameters. In this study, a conceptual BN model is developed to estimate the probability of outburst intensities in the laboratory. A total of 2,000 groups of random samples based on physical experiments are produced by trace linear or 3D interpolation for these factors and outbursts, respectively. A total of 1,800 groups of random samples are used to construct the BNs while the other 200 random samples are used for validation purposes. The proposed BN model estimates the intensity of the outburst with a root-mean-square deviation (RMSD) (3.8) and a linear correlation (0.94) compared with validation values of 200 random samples. [Received: December 2, 2014; Accepted: May 2, 2015]

Online publication date: Fri, 02-Sep-2016

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Oil, Gas and Coal Technology (IJOGCT):
Login with your Inderscience username and password:

    Username:        Password:         

Forgotten your password?


Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com