摘要
起重机小车工作中会产生机械震动而引发拖链电缆的脱槽,因此本文基于视觉实时监控拖链运行状态进行研发监控系统,通过在小车上安装工业级防抖摄像头,基于卷积神经网络的MaskR-CNN实例分割算法,实时监控拖链运行状态,自主开发检测脱轨算法,识别拖链算法,通过边缘点提取中心线,根据中心线来获得左右面积之比,若结果超过给定阈值,即为拖链脱轨。一旦检测到拖链出现脱槽情况,会立刻发出警示信号并通知系统立刻停车。经过实验结果表明,该系统能够有效保证系统免受破坏,避免造成长时间停机故障。
The crane trolley work will produce mechanical vibration and trigger the derailment of the drag chain cable,so this paper develops a monitoring system based on visual real-time monitoring of the running state of the drag chain,through the installation of industrial-grade anti-shake camera on the trolley,Mask R-CNN instance segmentation algorithm based on convolutional neural network,real-time monitoring of the drag chain operation status,self-developed algorithm for detecting derailment,identification of drag chain algorithm,extraction by edge point Center line,according to the center line to get the ratio of left and right area,if the result exceeds the given threshold,it is the drag chain derailment.Once the derailment of the drag chain is detected,a warning signal will be issued immediately and the system will be notified to stop immediately.After research and experimental results show that the system can effectively ensure the system from damage and avoid causing long downtime failure.
作者
周佳明
韩冲
周嘉钰
蔡桢杰
郝聚涛
ZHOU Jiaming;HAN Chong;ZHOU Jiayu;CAI Zhenjie;HAO Jutao(College of Electronics and Information,Shanghai Dianji University,Shanghai,China,201306)
出处
《福建电脑》
2023年第2期73-77,共5页
Journal of Fujian Computer
基金
上海市大学生创新创业训练项目(No.202111458032)资助。
关键词
供电拖链
防脱槽
预警系统
实例分割算法
Power Supply Towline
Anti-dislocation
Early Warning System
Mask R-CNN
作者简介
通信作者:周佳明,男,2001年生,主要研究领域为软件相关。E-mail:2824492695@qq.com;韩冲,男,2000年生,主要研究领域为后端开发、图像处理。E-mail:1530450010@qq.com;周嘉钰,女,2002年生,主要研究领域为软件工程。E-mail:2857255607@qq.com;蔡桢杰,男,2001年生,主要研究领域为机器学习、web安全。E-mail:44maker_01@proton.me;郝聚涛,男,1976年生,主要研究领域为数据挖掘。E-mail:haojt@sdju.edu.cn。
