Using swarm intelligence optimization algorithms to predict the height of fractured water-conducting zone

The accurate calculation of the height fractured water-conducting zone (FWCZ) is great significance for mine optimization design, water disaster prevention, and safety production coal mines. In this article, a height-prediction model FWCZ based on extreme learning machine (ELM) proposed. To address issues low prediction accuracy challenging parameter optimization, we optimized ELM using gray-wolf algorithm (GOA), whale (WOA), salp (SOA). These algorithms mitigate slow convergence, poor stability, local optimality associated with traditional neural networks. mining depth, height, overburden strata structure, working face length, seam dip angle are selected as main controlling factors FWCZ. A total 42 fields-measured samples collected divided into 2 subsets training validating ratio 36/6. capability GOA-ELM, WOA-ELM, SOA-ELM models evaluated compared, results show that three compared model. GOA WOA similar, while better than other two models, relative errors test sets all less 10%. Therefore, finally applied to predict formed after No.15 in Xinjian Coal Mine. Finally, verified measured data from borehole television detection instrument, which showed good consistency. This provides further evidence effectiveness swarm intelligence predicting