Transforming post-mining area into expressway site by stability evaluation with clustering method: A case study

Abstract

ABSTRACT Mining subsidence, as one of the most common geological hazards, bringing a massive threat to the new infrastructure in such a post-mining area. To evaluate the stability of a post-mining area transforming into an expressway site, a hybrid model based on the kernel principal components analysis (KPCA) and the fuzzy C-means clustering algorithm (FCM) was proposed by multi-attribute geo-mining variables. This study attempt to cover the inaccuracy clustering gap of conventional approaches with geo-mining stability influencing factors (SIFs), the model was established to cluster the status of a post-mining area with 9 geo-mining SIFs through 134 mined-out working panels of 6 collieries underneath the proposed Jining expressway site in Shandong province, eastern China. Taking the results of ground residual deformation calculated by the probability integration method (PIM) as a validation set, the prediction accuracy of the proposed KPCA-FCM model and conventional clustering approaches was compared. The major experimental results indicate that the maximum residual subsidence in the study area is 96 mm, after KPCA dimensionality reduction, the first four principal components of cumulative contribution rate were 83.83%, the external performance indicator Rand Index (RI) of the KPCA-FCM algorithm is 0.86, consistent with the actual status of the stability in the post-mining area. Overall, the performance of the proposed approach demonstrates the effectiveness of the stability evaluation in the post-mining area with multi-geo-mining SIFs and it is feasible to transform the post-mining area into the proposed expressway site.