Research on mine safety situation forecast and early warning

Abstract

Based on the Internet of Things technology, obtaining the mine safety big data and making full use of the data are helpful to realize the forecast and early warning of mine safety situation. Taking the gas explosion accident as an example, by analyzing the cause of the accident, a mine safety situation evaluation index system is constructed, and each evaluation index is quantified. Based on the long and short-term memory(LSTM) network and the Bayesian network, a mine safety situation forecast model is proposed. According to the mine safety monitoring data, the mine safety situation evaluation index forecast values are obtained through the LSTM. The risk probability of mine safety accidents is inferred from Bayesian networks based on the evaluation index forecast values to obtain mine safety situation forecast. Based on the safety situation forecast results, an early warning mechanism is established. 4 warning levels and response departments are classified according to the warning situation, and corresponding early warning measures are established. An inversion of a gas explosion accident in a coal mine is used as an example, and the results show that the forecast results of mine safety situation based on LSTM and Bayesian network are consistent with the actual situation.

Keywords

• mine internet of things; mine big data; mine safety situation; safety situation forecast and early warning; gas explosion accident; bayesian network; long and short-term memory network