A Prediction Model for Top-Coal Drawing Capability in Steep Seams Based on PCA–GRNN

To improve prediction accuracy of top-coal drawing capability in steep seams, principal component analysis (PCA) and the general regression neural network (GRNN) are combined (PCA–GRNN model) to predict seams. Nine commonly used influencing factors selected establish a predictive index system for The PCA is eliminate correlation reduce dimensions various indices, thus obtaining three linearly uncorrelated components (PCs) y 1 , id="M2"> 2 id="M3"> 3 which form input vectors GRNN. In this way, that most affect seams found be floor flatness, dip angle coal seam, hardness seam. results show PCA–GRNN model outperforms GRNN random forest models results, indicates improves model. It feasible by combining or even integrating different analytical into one. proposed can