基于视觉识别的煤矿输送带AI智能监测系统研究  

Research on Al intelligent detection system of coal mine conveyor belt based on visual recognition

作　　者：王蒙 刘鹏 王帅 WANG Meng;LIU Peng;WANG Shuai(Halagou Coal Mine,CHN Energy Shendong Coal,Shenmu 719300,Shaanxi,China)

机构地区：[1]国家能源集团神东煤炭哈拉沟煤矿,陕西神木719300

出　　处：《矿山机械》2025年第4期24-29,共6页Mining & Processing Equipment

摘　　要：介绍了煤矿输送带AI智能监测系统的构建与应用,阐述了系统的研究背景、技术路线、设备配置、试验过程及应用效果。系统集成了人工智能图像分析技术与长距离定位技术,实现对输送带表面损伤及接扣异常的精准识别与定位。矿用隔爆兼本安型相机实时采集输送带图像,并利用其高帧率、低曝光的特点,在物体高速运动状态下依然能取得高清图像。取得的图像由工业网络传回服务器进行图像分析,结合定位算法实现对异常告警的跟踪与定位,同时系统具备定位停机功能。The construction and application of the AI intelligent monitoring system for coal mine conveyor belts were introduced,and then the research background,technical route,equipment configuration,test process and application effect of the system were described.The system integrated artificial intelligence image analysis technology and long-distance positioning technology to achieve accurate identification and positioning of conveyor belt surface damage and abnormal fastening.In addition,the mine flameproof and intrinsically safe camera was used to collect the image of the conveyor belt in real time,and its high frame rate and low exposure characteristics was used to obtain high-definition images in the high-speed motion state of the object.Then,the obtained image was transmitted back to the server for image analysis by the industrial network,and the tracking and positioning of the abnormal alarm was realized by combining the positioning algorithm.At the same time,the system had the function of shutdown positioning.

关 键 词：煤矿输送带 AI智能监测 图像分析 长距离定位 实时采集 

分 类 号：TD528[矿业工程—矿山机电] TP18[自动化与计算机技术—控制理论与控制工程]