Research On Railway Accident Data Forecasting Method Based On The Correlation Analysis

For railway transportation,production safety is the first priority.Prevent the happening of the railway traffic accident,to improve the efficiency of railway transportation,improve the transportation of railway system is stable and has larger significance.Study the law of railway accident,need the help of historical data,in-depth excavation there is causality.From railway operational indicators,both passenger turnover and freight turnover,China railway completed quota is the world’s highest,and the railway point,line length,wide radiation characteristics and determines the probability of the accident is bound to increase.In the process of railway transport in the long run,produce the massive amounts of data,there are plenty of data in terms of traffic accident can be used to analysis and research,scientific research,these data can provide data support for the decision-making and command,is of great significance to the safe and stable operation of railway.In this paper,based on the actual data,quantitative study all aspects of the railway accident.First of all,on the influence factors of railway accident data statistical analysis,to facilitate subsequent analysis of the data mining,find the rule,there were a dominant lays the foundation for later research.Secondly,introduces the concept of the grey correlation analysis,expounds its basic theory and basic characteristics,the calculation steps were introduced.Then calculated of grey correlation degree of different types of accidents,accident are obtained,we analyzed the factors affecting the largest railway accident.Combined with the correlation analysis,analyze the correlation of different types of accidents,and put forward the corresponding analysis results.The introduction of the concept of combination forecast,in the difficulty is to determine the weighting coefficient of combination forecast,and test the effectiveness of the combination forecast model to analysis by error evaluation.The IOWA operator is introduced to set up the combined forecasting model based on the IOWA operator,to analyze its expression,lists the calculation steps,based on the grey relation analysis of IOWA operator combination forecast model.The railway accident forecast comparison on the foundation of analysis.Selected as single forecast model GM(1,1)model,combined with real data,railway accident prediction for seven types of accidents and the predictive value of virtual in 12 months,use the combination forecast model based on the IOWA operator,using the combination forecast model based on grey correlation degree of the IOWA operator,seven class separately predict accident in 12 months and virtual data.Compare three kinds of prediction model forecast analysis,and analyzes the error,the results showed that the IOWA operator combination forecast model based on grey correlation results from both with the actual value of the degree of fitting,or from the point of error precision,the effect is the best,so as to establish the model of effective,can be used in the railway accident prediction,priorities for prevention of railway accident has a certain guiding role.The last is the conclusion and prospect,summarize the point of view,for the next step to study and put forward the corresponding problems.