Integrated three-dimensional visualization and soft-sensing system for underground paste backfilling

Cemented paste backfilling (CPB) has attracted worldwide attention because of its advantages in the processing of waste tailings and underground goafs. A problem of the underground CPB procedure that has received little attention is the shortage of devices or techniques for monitoring the real-time backfilling process accurately due to the harsh underground environment. Imprecise information of the backfilling progress will mislead the operators to make improper decisions, such as not pausing the transport system promptly or tuning the ratio of cement to tailings incorrectly. This study proposes a 3D visualization and soft-sensing system for the cemented paste backfilling process. By employing a GeoSLAM ZEB-HORIZON laser scanner to scan the 3D model of the underground goaf, the system estimates the backfilling height in real-time according to the accumulated volume of filled paste and simulates the backfilling of the underground goaf on web pages. In a real Copper Mine, we examined the developed system and soft-sensing technique in goafs and found that the relative error of estimated backfilling height is under 10%, which is acceptable in real production. The results indicate that the developed backfilling visualization and the soft-sensing techniques provide significant guidance for production management in underground backfilling.

Automated summary

Cemented paste backfilling is an effective method for processing waste tailings and underground goafs. However, there is a lack of devices or techniques for accurately monitoring the real-time backfilling process in harsh underground environments. This study proposes a 3D visualization and soft-sensing system that estimates the backfilling height in real-time using a laser scanner and simulates the process on web pages. The developed system and technique provide significant guidance for production management in underground backfilling.