Research On The Prediction And Early-warning Models Of Gas Explosion Disaster In Coal Mine Internet Of Things Environment

The prediction and early warning technologies of gas disaster are the effective method to prevent the occurrence of serious accidents and improve the coal mine production safety.The system of coal mine internet of things is that applying the internet of things technology into coal mine production.And it can realize flexible,dynamic and real-time acquisition of various types of coal mine data,and effectively support the research on the information processing and recognition technology in the application layer.Thus can realize the goal of real-time information,intelligent control,accuracy prediction and early-warning,and effectively reduce the number of coal mine accident death,and improve coal mine production safety.In this paper,the prediction and early-warning models of gas explosion disaster are studied under the environment of coal mine internet of things.The specific research contents of this paper are as follows:Firstly,The technical status and system application status of the coal mine internet of things and the research status of gas explosion disaster prediction and warning technology are expounded.The application of the coal mine internet of things can solve the problems of single data type and poor real time in the traditional gas monitoring system,and realize the real-time acquisition of many types of data in underground coal mine.This paper introduces the application of coal mine internet of things in the shanbula coal mine and the information processing and recognition technology in detail.Secondly,in this paper the gas explosion disaster prediction technology is researched,and the BP neural network prediction model based on PSO-MEA hybrid optimization algorithm is established.The proposed PSO-MEA hybrid optimization algorithm is used to optimize the weights and thresholds of the BP neural network model.Because of the combination of particle swarm optimization(Particle Swarm Optimization,PSO)and mind evolutionary algorithm(Mind Evolutionary Algorithm,MEA),the accuracy of prediction model is improved.Thirdly,in this paper the gas explosion disaster early-warning technology is researched.This paper established a nonlinear fuzzy comprehensive evaluation model,and this model is used to evaluate the degree of gas explosion disaster,and then the early-warning model of gas explosion disaster is formed.The model can realize the accurate evaluation of the risk level and improve the accuracy,at the same time,it has the advantages of wide application,full extraction and utilization of existing information,and small loss of information.Finally,the prediction and early warning models of gas explosion disaster are simulated by MATLAB software,through the real-time data of Shanbula coal mine.By comparing with other models,it is found that the maximum average output error of the prediction model of gas explosion disaster is 0.0174,and the minimum is 0.0078.Comparing with PSO-BP model and MEA-BP model,the minimum output precision is improved by 3.59% and 3.27%,respectively.The early-warning model of gas explosion disaster achieves accurate and precise classification.