采煤工作面瓦斯涌出量的聚类预测

Cluster analysis and prediction of the amount of gas emitted from coal face

以预测回采工作面瓦斯涌出量为研究目的,应用模糊C均值聚类算法对回采工作面瓦斯涌出资料进行聚类分析,得到各样本的类别、聚类中心和样本对于各类的隶属度;根据瓦斯涌出量值建立分类标准,验证了模糊C均值聚类算法在瓦斯涌出量分类方面的实用性;根据瓦斯涌出量的影响因素和聚类中心,应用灰色关联分析法建立回采工作面瓦斯涌出量的归类预测模型,对待测样本进行预测,实现了瓦斯涌出量的聚类预测;最后,用实例论证了该方法的可行性和有效性.

The research purpose is to predict the amount of gas emitted from coal face. The gas emission data are analyzed using the fuzzy C-mean cluster algorithm. The sample classification, Cluster centers and membership degrees of samples are obtained. The classification criterion of gas emission is established to validate the veracity of fuzzy C-mean cluster algorithm. According to the influencing factors of gas emission and cluster centers, categorization prediction model for gas emission is founded by using grey incidence analysis method, and is applied to predict testing samples. Based on the combination of the fuzzy C-mean cluster algorithm and the categorization prediction model, the categorization prediction of gas emission is achieved. Finally, the method is applied to practical examples. The application shows that the method is feasible to predicting the amount of gas emitted from coal face.