Coal Safety Monitoring System Based On Deep Learning Multi-Sensor Data Fusion Early Warning Model

According to statistics,many mine accidents caused by gas leaks occur every year in China,which not only threatens the life safety of miners,but also causes serious economic losses.Therefore,mine environmental safety is the top priority of the safety management of the coal industry.It is necessary to accurately describe the environmental status of the mine in order to present more intuitive judgments to the staff and reduce unnecessary safety hazards.This paper analyzes the causes of common coal safety accidents and simplifies the mine environment state into three states.Appropriate environmental parameters,methane,oxygen and carbon monoxide,were chosen.In order to realize the identification of the environmental status of the mine,reduce the ambiguity of the environmental parameters,and increase the reliability of the monitoring system.In this paper,good expression and DBN model feature classification ability,fusion method based on the DBN model,using the state of the environment with the environmental parameters on the label DBN model training and learning multi-sensor data fusion,to find the relationship between the environmental parameters and environmental conditions,so that the monitoring system can describe the actual state of the mine environment,to provide accurate determination of staff.In this paper,the corresponding sensor is designed to realize the collection of environmental parameters,and the transmission substation is designed to summarize the environmental parameters collected by the sensor,and then the environmental parameters are packaged and uploaded to the upper computer software through Ethernet.The upper computer software uses the environment parameter with the environmental status label as the input of the DBN model,and calls the DBN algorithm generated by MATLAB to train the DBN model to obtain the multi-sensor data fusion model.Finally,the upper computer software input the real-time acquired environmental parameters into the data fusion model completed by training to identify the mine environment status,and realize the monitoring and description of the mine environment by the coal mine safety monitoring system,and verify the accuracy and reliability of the system through testing.