Gong-kuang zidonghua (Jun 2021)

Top coal thickness detection method for intelligent fully-mechanized working face

  • YANG Xiuyu1,
  • LIU Shuai2,3,
  • LIU Qing2,3,
  • YANG Qingxiang1

DOI
https://doi.org/10.13272/j.issn.1671-251x.2020080059
Journal volume & issue
Vol. 47, no. 6
pp. 79 – 83

Abstract

Read online

Detecting the thickness of the top coal in a fully mechanized working face in advance can provide a basis for precise control of the top coal caving, which is beneficial to achieving a balance between the coal caving recovery rate and coal quality. By analyzing the principle of ground-penetrating radar detecting coal-rock interface and coal seam thickness, a ground-penetrating radar device for detecting top coal thickness is designed, and an intelligent method for detecting the thickness of top coal in fully mechanized working face based on ground penetrating radar is further proposed. The radar pulse wave is transmitted and received by the ground-penetrating radar device, and the received signal is amplified, sampled and integrated to form a radar frame, which is transmitted to the control unit of the unmanned coal mining machine in real time by WiFi and finally to the console of the central control room. The coal-rock interface extraction software processes and analyzes the reflected signal waveform and gray-scale image, and determines the position of the coal-rock interface according to the maximum and minimum amplitude of the reflected signal. The method calculates the coal seam thickness by the time difference between the position of maximum or minimum amplitude and the starting point of radar pulse emission. The method is tested in the 12309 fully mechanized working face of Wangjialing Coal Mine. And the results show that the top coal thickness detection result interpreted by radar reflection wave gray-scale image at a certain place is 3.383 m, and the error of the actual value (3.16 m) detected manually is 7%. The maximum thickness of the top coal that can be detected is 5 m, and the maximum detection error does not exceed 10%. The performance meets the actual detection requirements.

Keywords