基于机器视觉的矿井煤矸分拣装置研制  

Development of Coal Gangue Separation Device Based on Machine Vision for Mines

作　　者：胡平 李新 吕晨辉 赵安新[2] 李学文[2] 杨金桥 Hu Ping;Li Xin;Lyu Chenhui;Zhao Anxin;Li Xuewen;Yang Jinqiao(Inner Mongolia Shuangxin Mining Co.,Ltd.,Ordos 017000,China;Xi'an University of Science and Technology,Xi'an 710054,China)

机构地区：[1]内蒙古双欣矿业有限公司,内蒙古鄂尔多斯017000 [2]西安科技大学,西安710054

出　　处：《煤矿机械》2022年第12期211-213,共3页Coal Mine Machinery

基　　金：智慧矿山开放基金项目(2019LH01);西安市科技计划资助项目(2020KJRC0066);陕西省重点研发计划(2021SF-479)。

摘　　要：针对煤炭生产过程中煤矸分拣靠人工分拣、机器分拣不够准确的问题,选用机器视觉智能AI识别及五联机械手,配合漏抓、错抓判别控制技术,采用两级目标检测网络,利用执行机构XYZ运动模拟人工抓取,研制了矿井煤矸分拣装置。通过在某煤矿206带式输送机上增加该装置,实现了智能煤矸分选,识别率可达95%以上,比人工分拣的捡矸率至少提高40%。Aiming at the problem that coal gangue separation depends on manual sorting and machine sorting is not accurate enough in the coal production process, used machine visual AI intelligent recognition and five-link manipulator, cooperated with the control technology of missed catch and wrong catch,adopted two-level target detection network, used XYZ motion to simulate manual grabbing, developed a coal gangue separation device for mines. Realized the intelligent coal gangue spearation by adding this device to the 206 belt conveyor in a coal mine. The recognition rate can reach more than 95%, the gangue picking rate is at least 40% higher than the manual sorting.

关 键 词：机器视觉 煤矸分拣 卷积神经网络 仿生机械手 

分 类 号：TD67[矿业工程—矿山机电]