煤与瓦斯突出预报的理想灰贴近函数聚类关联分析

Ideal gray function cluster correlated analysis on forecast for the coal and gas outburst

从预报煤与瓦斯突出所获取的数据信息知,其信息多数为灰色特征。为突出其灰色特性,本文采用理想灰贴近函数聚类关联分析预测预报煤与瓦斯突出的危险程度,并将此方法同模糊聚类预测预报法和模糊聚类关联分析预报法及模糊聚类相似分析预报法进行了对比分析。由对比分析可知,采用4种数学方法预测预报煤与瓦斯是否突出,给出的结论是一致的,只是预测预报是否发生的可能程度略有差别。其中关联分析比相似分析计算过程较简单,且相似分析在确定编号时,有时受截值λ取值影响。尽管如此,但不影响预测预报结果。所以,不论是采用模糊聚类,还是灰色聚类数学方法预报煤与瓦斯突出均是可行的,可在实践中应用。

To highlight the gray factors of the main data information gained from the forecasting coal and gas outburst, a method with ideal gray function and cluster correlated analysis is used to forecast and predict the danger of coal and gas oubturst in this paper. And a comparison is presented among the prediction methods of fuzzy cluster, fuzzy correlated analysis, fuzzy cluster similar analysis and this prediction method. The final analysis shows that the same result comes from the four mathematical prediction methods and the mathematical methods of both fuzzy cluster analysis and gray cluster analysis can be used in forecasting and predicting coal and gas outburst, while the possibility of occurring is a little different. In addition, the precision of correlated analysis is lower than that of similar analysis, and the primary reason is that correlated analysis is confined by ditinguishing coefficient, and when defining number correlated analysis may be affected by 'λ' sometimes. Although it is not influence the effect of forecast and predict. At the same time, both fuzzy cluster and gray cluster methods that are used to forecast coal and gas outburst are feasible and can be applied in practice.