Development and application of a visual intelligent supervision system for coal flow transportation

SUN Xiaohu; XU Hao; GAO Jie; ZHANG Beining; HU Bing; YIN Long; WU Zhefeng; GAO Shuchen; ZHAO Xinwen

doi:10.11799/ce202511014

Abstract：To address the frequent failures and operational disruptions of underground belt conveyors in coal mines,which operate in humid,high-temperature,and dusty environments,leading to increased downtime and costs,this paper proposes an intelligent monitoring and control system based on a cloud-edge collaborative architecture and a continual learning AI framework.By integrating computer vision,equipment control,and big data analytics,algorithm models for coal flow anomaly detection were developed,resulting in a visual intelligent monitoring and management system.This system enables intelligent speed regulation of conveyors based on real-time coal volume,and achieves closed-loop management covering damage tracking,anomaly warning,and incident response.Industrial deployment at mines such as Hetaoyu Coal Mine demonstrates the system's effectiveness:belt speed measurement error is within±2%,coal volume statistical error is less than±5%,conveyor idle rate is reduced by 30%,annual electricity savings reach approximately 150000 kWh/km,and roller replacement cycles are extended by 40%.Furthermore,the system reduces daily inspection manpower by 10 person-times,cuts abnormal downtime by 15 minutes per day,and generates annual direct economic benefits of 2 million yuan and indirect benefits of 10 million yuan.This system significantly enhances the reliability,economy,and intellig and intelligence of belt conveyor operations,providing key technical support for efficient and green transportation in coal mines.