基于机械臂的煤块粒度自动识别系统研究  

作　　者：孙宇晨 陈雨轩 王聪[1] 陈勃帆 孙宇 刘同冈[1] 

机构地区：[1]中国矿业大学,江苏徐州221116

出　　处：《科技创新与应用》2023年第25期33-37,共5页Technology Innovation and Application

基　　金：江苏省省级大学生创新训练项目基金资助课题(202210290151Y)。

摘　　要：煤炭生产过程中,煤块破碎的粒度大小影响着产品的质量,为使煤块粒度符合不同使用需求,应当对煤块粒度分布进行识别与统计。为提高煤块粒度的识别效率,设计基于机械臂的煤块粒度自动识别系统,采用随机取样的统计方法,使用机械臂铲取煤样并摊平,利用机器视觉技术识别并用OpenCV中的分水岭算法统计煤块粒度大小,实时掌握煤块粒度情况。测试结果表明,系统对煤块粒度识别的误差率为7.1%,能够满足现场生产需求。In the process of coal production,the particle size of coal lump affects the quality of products.In order to make the particle size of coal lump meet different requirements,the particle size distribution of coal lump should be identified and counted;in order to improve the recognition efficiency of coal lump size,the designed coal lump size automatic identification system based on mechanical arm adopts the statistical method of random sampling,uses the mechanical arm to shovel and flatten the coal sample,uses the machine vision technology to identify and use the watershed algorithm in OpenCV to calculate the coal lump particle size,and grasps the coal lump particle size in real time.The test results show that the error rate of coal particle size recognition of the system is 7.1%,which can meet the needs of on-site production.

关 键 词：机械臂 煤块识别 机器视觉 OPENCV 分水岭算法 

分 类 号：TP241[自动化与计算机技术—检测技术与自动化装置]