Statistics Analysis And Forecasting Of Coal Mine Production Safety Accidents In Our Country

Collect the coal mine production safety accidents data from 2001 to 2015 which was released by the query system from State Administration of Coal Mine Safety in China.Analyze the features of the accident in the ways of Statistical Graph,Pareto Chart,Management of Figure,Test of Significance,Principal Component Analysis and Hierarchical Cluster from the aspects of year,quarter,month,kind,territory and the death rate.It combined the strongly discrete data,the little data quantity,and nonlinear etc.According to the different.It provide reference for scientific prevention and control measures of coal mine safety production in China.According to the characteristics with complex condition of safety production in coal mine in our country,choose the ARIMA model in the time series prediction method and exponential smoothing prediction method to predict the death toll and number of coal mine production safety accident.Choose the support vector regression machine prediction method to predict the total loss of a large state-owned coal mine production safety accident.These methods in the prediction of coal mine security situation in the future in our country,reflects the higher accuracy,and provides support to decision making in order to prevent accident.