Research On Coal Mine Gas Dynamic Prediction Based On Intelligent Information Processing

Coal mine is an important strategic resource of energy; it plays a very important role in national economic development. In the process of coal mining, mine safety issues are frequent. Gas disasters are very common accident; it caused a large number of casualties and economic losses of property. Therefore, gas disasters seriously affected the coal industry's sustainable development. In order to improve the prediction of gas safety levels and reduce harmful gas disaster, gas dynamic study of intelligent forecasting methods and models are essential. An adaptive and better accurate gas dynamic prediction method is researched; and gas disasters have high gas emission, gas explosion, coal and gas outburst characteristics, combined with intelligent information processing technology, the gas emission and gas concentration and coal and gas outburst and other aspects of gas disaster's forecasting and analysis methods are thoroughly investigated. Research work mainly includes the following aspects:1) Gas emission and gas concentration factors and coal and gas outburst mechanism and influencing factors are analyzed, so, the prediction methods are divided into numerical calculation method and classification calculation method. By analyzing these three gas aspects, gas emission and gas concentration can be predicted by numerical calculation method, coal and gas outburst can be predicted by classification calculation method. And information generation algorithm and gas dynamic prediction model are researched.2) New intelligent predictive methods are proposed(Opposite Degree-Numerical Computation algorithm and Opposite Degree- Classification Computation algorithm). The methods are based on the degree of semantic opposites, neural networks, genetic algorithms and cluster analysis, and they can be applied in numerical prediction and classification prediction.3) Information Generation algorithm is proposed. The algorithm is divided into two sentence template, namely Numerical Computation Sentence Template(NCST), Classification Computation Sentence Template(CCST). IG can combine with OD related algorithms and prediction model and early warning for coal mine.4) OD algorithm and IG algorithm for the prediction of gas has been studied. The study includes: gas emission and gas concentration is predicted based on the OD-NC and IG, coal and gas outburst and gas security status is predicted based on the OD-CC and IG. Among them, the gas security state prediction has a very close relationship to gas emission, gas concentration, coal and gas outburst. Predicting gas security status is an important complement to the results of the algorithm. By comparing the prediction results of different intelligent algorithms, the prediction accuracy of the algorithm is verified; by using expert scoring method, information generation algorithm evaluation results in methane prediction is explained.5) By using ASP.NET + C # and Access 2007 technology, intelligent gas prediction model was realized. The model incorporates the actual situation in Anhui Guobei coal mine; gas concentration in the sample application is predicted based on opposite degree and information generation. After verification instance, further evidence of the reliability and intelligence of the algorithm and model are shown. And the information generation result is consistent with the actual situation.In summary, the OD-NC and OD-CC have good generalization ability and fitness and high prediction accuracy; OD-NC and OD-CC and IG algorithm are feasible and effective. The studies of gas dynamic forecasting methods and models based on intelligent information processing technology for coal mine safety supervision and have a positive effect. Its forecasting methods and models help prevent and reduce coal mine gas disasters, but also conducive to provide program support and theoretical reference for gas safety related research.