28 August 2019 in International Journal of Distributed Sensor Networks
Longwall mining is a productive mining method that has been widely used in China, the United States, Australia, and Europe. However, due to the subsidence caused by coal mining activities, the phenomenon is complex, longwall mining brings the overburden movement issues accompany with the coal recovery. In subsidence trough, the resulting strata and surface ground movements are regarded as largely contemporaneous with mining, producing more or less direct effects of natural ecology. Ground penetrating radar has been widely used in geological hazard detection due to its high precision, but it is rarely employed in underground measuring the overlying strata movement above the longwall panel. In this article, there are five close-multiple coal seams (5#, 7#, 8#, 9#, 12# seams) within 70 m distance to be monitored in Qian Jiaying coal mine of Kailuan Coal Company in Hebei province. The 7#, 8#, 9#, 12# coal seams were already excavated in sequence. So the 5# coal seam is overmining above the gobs of four coal seams at the top. The ZTR12-series ground penetrating radar equipment is used to detect the development of fractures under the floor of the 5# coal seam. The ZTR12-series ground penetrating radar system has the capability of large depth detection and explosion proof to adapt to the working environment of underground coal mines and can realize fine detection and accurate identification. The maximum detection distance of reflection method can reach 50 m. The measurement results show that the 5# coal seam is in the fracture zone above the gobs of four coal seams, and the caved zone of lower coal seam develops to fine sandstone of the 5# coal seam floor. The ground penetrating radar has shown much promise in the detection of overlying strata movement.
7 August 2008 in Journal of Coal Science and Engineering (China)
1 January 2021 in Advances in Civil Engineering
The neural network algorithm is a small sample machine learning method built on the statistical learning theory and the lowest structural risk principle. Classical neural network algorithms mainly aim at solving two‐classification problems, making it infeasible for multiclassification problems encountered in engineering practice. According to the main factors affecting sand liquefaction, a sand liquefaction discriminant model based on a clustering‐binary tree multiclass neural network algorithm is established using the class distance idea in cluster analysis. The model can establish the nonlinear relationship between sand liquefaction and various influencing factors by learning limited samples. The research results show that the hierarchical structure based on the clustering‐binary tree neural network algorithm is reasonable, and the sand liquefaction level can be categorized accurately.