Cause analysis of rockburst disasters based on FP-growth algorithm and Bayesian network
OUYANG Zhenhua
XIAO Manman
LIU Jian
XU Qianhai
JU Chengrun
Abstract：To overcome the limitations of traditional methods in analyzing the multi-factor coupling disaster of rock burst,accurately decipher the causal logic of this hazard,and achieve proactive risk management,this study utilizes 56 authoritative accident investigation reports from 2001 to 2024 as samples.The FP-growth al-gorithm is employed to mine unstructured text,extracting high-frequency co-occurrence causes through a three-dimensional filtering mechanism based on support,confidence,and lift.Subsequently,the mining results are integrated with domain knowledge to construct a Bayesian network topology with prior-posterior dynamic updating capability,enabling probabilistic reasoning of the causal chain and risk inversion.The study reveals the nonlinear disaster patterns arising from the coupling of "geology-mining-management" factors,quantita-tively identifies the dominant role of geological factors such as coal-rock impact propensity,geological struc-tures,and mining depth,as well as the key triggering effects of mining factors like excavation disturbances and coal pillars.It also clarifies the risk amplification effect of management factors,such as insufficient safety awareness.Based on the quantitative analysis results,targeted proactive control measures for rock burst risk are proposed,providing a scientific basis for coal mining enterprises to optimize mining plans and precisely deploy pressure relief monitoring projects.This approach can facilitate the transition of rock burst prevention and con-trol from experience-driven to data-knowledge collaborative driven.
Keywords：rock burstdisaster-causing factorsFP-growth algorithmBayesian networkassociation rela-tionship
Publication Date：2026-02-28
Online Publishing Date：2026-03-12(First online date of this platform, not the publication date of the document)
Pages：11( 106-116 )