Search

Based on database of collected gas monitoring information of coal mine monitoring system in coal mine production process,the paper put forward a scheme of using clustering analysis method based on DBSCAN algorithm to mining characteristics of gas accident information.It pointed out shortage of DBSCAN algorithm in practical application,proposed an improved scheme of DBSCAN algorithm based on data partition,and introduced the concrete improving method.The result of applying the clustering analysis method based on the improved DBSCAN algorithm to make clustering analysis and extract of information feature of gas monitoring data of coal mine showed that the method is effective.At last,the paper pointed out the farther research direction of the method.

For the purpose of achieving more accurate and reliable gas emission dynamic prediction through effective analysis of gas measuring data in mines, this paper put forward a method that use ant colony clustering to optimize El-man neural network. Ant colony clustering algorithm was merged with Elman neural network to optimize weight and threshold. The model of gas emission quantity prediction was established by ACC-ENN algorithm, with the historical data of mine actual monitoring to experiment and analysis. The results show that the Elman neural network model optimized by ant colony clustering than other prediction model has better generalization ability and higher precision of prediction, to realize the dynamic forecast of gas emission effectively. [ABSTRACT FROM AUTHOR]