基于PCA-RBF网络的煤与瓦斯突出强度预测.

In order to increase the accuracy of coal and gas outburst intensity prediction, this paper proposed that principal component analysis (PCA) can be adopted to reduce the correlation between variables and radial basis function network which was combined to predict coal and gas outburst intensity. Taking a coal mine as the object of research, the principal components of influencing factors of coal and gas outburst intensity were extracted. And three principal components with contribution rate of accumulated variance which was more than 85% were selected and used to replace the six original factors. The principal components were regarded as the input parameters of radial basis function network and PCA-RBF network prediction model was built. The results showed that the average relative error of PCA-RBF network prediction model was 5.55%, and it was accord with the requirements of coal and gas outbursts prediction. [ABSTRACT FROM AUTHOR]