Optimize the early warning time of coal and gas outburst by multi-source information fusion method during the tunneling process

Abstract

Accurate and advanced early warning of coal and gas outburst is an important means to ensure the process safety of coal mining. It is a difficult problem to realize the reliable early warning of coal and gas outburst because the cause of the complex causes and disaster-causing mechanism has not been clearly defined. At present, the traditional early warning models have some problems, such as low degree of information fusion, non-dynamic decision-making, and short early warning time. To optimize the early warning time and improve the process safety of tunneling, a new early warning model is proposed. This new model is a multi-source information fusion dynamic early warning model based on combining Autoregressive Integrated Moving Average (ARIMA) and the Transferable Belief Model (TBM). The ARIMA described the changes of different types of sensor data with spatial heterogeneity over time, and the TBM fused different types of sensor data and makes dynamic decisions to achieve early warning of Coal and gas outburst. Finally, Acoustic Emission (AE), Electromagnetic Resonance (EMR), and Gas concentration data of NO.12223 conveyor roadway of Jinjia coal mine from August 3, 2017, to August 7, 2017, were used as the experimental data set. The proposed model was applied to the NO.12223 conveyor roadway of Jinjia Coal Mine to verify the dynamic early warning for the risk of coal and gas outburst. The results showed that the gas gush out with tunneling is identified 1 and 38 min in advance. It was about 9 min earlier than the earliest response EMR signal obtained by the linear regression prediction model of a single indicator used in the original monitoring and early warning system. The research results are of practical significance to optimize the time of early warning of coal and gas outburst and improve process safety risk control in coal mine production.