基于自适应权重模糊C-均值聚类的瓦斯突出预测

 传统的模糊C-均值聚类在处理煤与瓦斯突出的预测时，由于其对初始聚类中心的过度依赖而导致预测结果准确率的降低。为了准确预测煤与瓦斯突出，提出了一种基于自适应权重模糊C-均值聚类的煤与瓦斯突出预测方法。该方法将瓦斯浓度相关影响因素作为特征空间中的样本，采用高斯距离比例表示权重，动态计算每个样本对于类的权重，对特征空间中的样本进行聚类分析预测，降低了算法对初始聚类中心的依赖。实例验证表明：所提出的方法具有较高的预测精度，具有较大的实用价值。

Prediction on Gas Outburst of Fuzzy C - Mean Value Cluster base on Self Adaptive Weighting

Due to Fuzzy C-Means clustering algorithm in dealing with outburst of coal and gas prediction rely heavily on randomly select C clustering centers, so made its accuracy of predicted results easy to fall into the local optimum states. Therefore, this paper proposed an improved Fuzzy C-Means clustering algorithm for outburst of coal and gas prediction based on self-adaptive weights. The new method made the related factors of gas concentration as the characteristic of the sample space ,expressed weight by using the Gaussian distance ratio, computed the weights for every data according to the current clustering state dynamic, predicted the clustering by using characteristic of the space and no more did rely on the initial clustering center. The experiments indicate that the Fuzzy C-Means clustering algorithm based on self-adaptive weights is an effective fuzzy clustering algorithm, has more robust and higher clustering accuracy. Through an engineering example, the practicality and accuracy of the method is tested and verified.