A Cold Rolling Mill Based On Data Mining Study On Prediction And Early Warning Of Accident Hidden Danger

With the development of China's economy,the security of enterprises has been paid more and more attention,and the level of security management of enterprises also determines whether enterprises can operate healthily and efficiently.As a pillar industry in China,the steel industry plays a vital role in economic development.Similarly,the rolling mills have complicated processes in the cold rolling process.There are many risk factors and uncertainties in the production process,and safety production accidents also occur from time to time..The traditional security management method is the "problem-based" management mode,which can only be qualitatively analyzed,and lacks simple and intuitive quantitative analysis means,which is very passive in enterprise security management.However,with the continuous development of information technology and the continuous popularization and application of data mining technology in various industries in recent years,it also provides new technical means and methods for hidden dangers investigation and prevention of production safety accidents.Taking a cold rolling enterprise as an example,this paper analyzes the hidden danger data of the factory from September to September 2018 in 2017,and studies the establishment of the forecasting and early warning model for the safety production accidents of the factory,guiding enterprises to carry out safety management specifically and effectively.The development of hidden danger investigation and management work is of great significance to improving the safety management level of enterprises.Firstly,the R language is used to analyze the hidden danger data of the enterprise.After the Chinese word segmentation,the word frequency of the hidden vocabulary is obtained.After the word frequency is counted,the visual word cloud map is drawn,and then the importance of the hidden vocabulary in each month is used.The word frequency-inverse document frequency(TF-IDF)converts the word frequency per month into a weight.Finally,using the grey prediction theory,the hidden danger prediction model is established and the accuracy test is carried out.The number of hidden vocabulary occurrences with the highest weight per month in 2019 is predicted and the generated trend of hidden dangers is analyzed.Through the prediction and early warning research on the safety hazard of the cold rolling mill,it has laid a solid foundation for the visualization of the safety production hazard investigation and the safety management of the enterprise.The conclusions reached are:(1)Obtaining the overall hidden dangers of the enterprise between September 2017 and December 1818,and showing the most frequent occurrence of “oil pollution” during this period through the word cloud.(2)Predicting the word frequency of 16 hidden vocabulary words in 2019,and generating a fitted graph to visually show the development trend of hidden dangers,reflecting the current occurrence of hidden dangers in the hidden danger vocabulary in the enterprise and the prediction of the future stage..(3)Establish a set of accident hidden danger prediction model and provide good technical support for the enterprise's safety production management work improvement.Through the establishment of a hidden danger prediction model for the cold rolling enterprise,the company can scientifically and intuitively display the actual situation of the current hidden dangers of the enterprise and the prediction of the hidden dangers in the next stage,so as to guide the enterprises to conduct targeted and targeted investigations on hidden dangers.Provide support to prevent accidents from occurring.