An intelligent approach using micro-seismic monitoring signal clustering and an optimized K-means model to guide the selection of support patterns in underground mines
Yunbo Tao, Qinli Zhang, Qiusong Chen, Chongchong Qi, Yikai Liu
Abstract
Determining the support patterns of underground mines is challenging due to the complex structure of the surrounding rock. To address the low accuracies inherent to empirical methods for determining support patterns and the low utilization efficiency of micro-seismic (MS) systems, this study proposes a method used MS signals to determine support patterns. The mining area of a copper mine was divided into 679 stopes, the three-dimensional coordinates of which were used to screen MS signals. A total of 2372 MS signals were considered and the sum of each MS feature in the stope was calculated. By combining K-means algorithms and meta-inspired intelligent optimization algorithms, an approach for MS signal-based support pattern selection was developed. It can be concluded that the optimal number of clusters for K-means analysis is 3, and the whale optimization algorithm K-means model has a better clustering performance. Cluster center sums from smallest to largest correspond to different support level, and stopes without MS signals do not need support. Field investigations and stope clustering confirm the utility of integrating MS signals for support selection, enhancing underground safety, and MS monitoring systems utilization. The proposed approach may be used to guide support strategies in mining areas.