Application of multidimensional structural equations in the emergency management of coal mine accidents

Zhang, Tianyue, Liu, Jiayu and Hong, Liang 2023. Application of multidimensional structural equations in the emergency management of coal mine accidents. Processes 11 (3) , 882. 10.3390/pr11030882

PDF - Published Version Available under License Creative Commons Attribution. Download (1MB)

Abstract

The use of coal as a source of energy is crucial for the growth of the national economy, but mining poses numerous risks and a potential for significant disasters. Coal mine safety is the prerequisite and guarantee for coal industry to achieve new industrialization and sustainable development. Therefore, it is crucial to predict a safety accident in the coal mine in advance. In order to facilitate the early warning of coal mine safety accidents, this study seeks to present a prediction model based on emergency management of safety accidents, which is a fusion model of principal component analysis (PCA) and long short-term memory neural network. According to the results, the correlation coefficients of risk identification and monitoring (a11), safety inspection and warning (a12), emergency planning and training (a13), material and technical support (a15), and macroenvironmental management (a21) were 0.718, 0.653, 0.628, 0.444, and 0.553, respectively, after the PCA dimensionality reduction process, demonstrating that the previous principal component analysis had a better effect. The absolute relative errors of each evaluation index of safety accident emergency management did not exceed the limit of 5%, including a15 and a21, whose values were 4.5% and −3.8%, while the relative errors of the remaining indicators were kept at a relatively low level. In conclusion, it is clear that the algorithm model suggested in this research improved the warning capabilities of safety accident emergency risk.

Item Type:	Article
Date Type:	Published Online
Status:	Published
Schools:	Business (Including Economics)
Additional Information:	License information from Publisher: LICENSE 1: URL: https://creativecommons.org/licenses/by/4.0/, Type: open-access
Publisher:	MDPI
Date of First Compliant Deposit:	6 April 2023
Date of Acceptance:	12 March 2023
Last Modified:	06 May 2023 09:36
URI:	https://orca.cardiff.ac.uk/id/eprint/158463

Actions (repository staff only)

Edit Item