A novel edge computing architecture for intelligent coal mining system
Zhe Bing，Xing Wang,Zhenliang Dong,Luobing Dong,Tao He
Abstract:The global coal industry, as an important basic industry, has strongly supported the stable and rapid development of the international economy and society. Although the level of automation in the context of coal mining has reached a very high degree, coal mine accidents such as roof collapse, side falling accidents, gas outburst potential, etc. continue to take place.Therefore, lots of intelligent systems are installed into collieries. For example, intelligent safety analysis technology represented by deep learning has been widely implemented in coal mine safety. The intelligent coal mining applications are always computing resource sensitive. They are usually deployed in the cloud centres that are located on the ground. As we all know, coal mining applications such as coal mine safety requires a comprehensive consideration of the mining,transportation, ventilation, hydrology, geology and other integrated factors. The transmission of all detecting data about these factors especially for multimedia from the underground face to the cloud computing centre is time-consuming.However, coal mine accidents always happen in a short span of time. This long transmission time is unacceptable for coal mine safety. In this paper, we propose a novel edge computing based intelligent processing architecture that integrates Internet of Things (IoT), fifth generation (5G), and Edge computing technologies for the coal mining intelligent system.Experiments are conducted on a deep learning based video fire prediction algorithm to prove the effectiveness of the architecture
Keywords Coal mine safety ;Intelligent processing architecture ; Edge computing ; Multimedia
Conclusion:In this paper, we propose a new coal mining safety moni-toring system architecture. It integrated the 5G wireless communication, edge computing, and intelligent tech-nologies. Using this architecture, intelligent coal mining safety preventing algorithm can be train based on the complete dataset and run in the edge device that is near to the mining face. This mechanism can minimize the pro-cessing delay of the intelligent algorithms. Finally, we conduct experiments to prove the effectiveness of our novel method using real colliery data. In the future, we will use this architecture in a colliery to improve its mining safety. We will also address the influence of the network transmission errors. Because these errors will make the computing unreliable.
Published journal: Wireless Networks ( IF 2.602 ) Pub Date : 2022-01-04