烧结机台车侧板裂缝检测与安全预警算法实现  被引量:1

Implementation of crack detection and safety early warning algorithm for side plate of sinter trolley

在线阅读下载全文

作  者:姜子欧 王月明[1] 陈甜甜 杨虎生 JIANG Ziou;WANG Yueming;CHEN Tiantian;YANG Husheng(School of Information Engineering,Inner Mongolia University of Science and Technology,Baotou 014010,China)

机构地区:[1]内蒙古科技大学信息工程学院,内蒙古包头014010

出  处:《冶金自动化》2023年第4期43-50,108,共9页Metallurgical Industry Automation

基  金:内蒙古自治区自然科学基金项目(2020MS06008);内蒙古自治区关键技术攻关项目(2021GG0045)。

摘  要:烧结机台车侧板在生产过程中容易产生裂缝,不仅会造成烧结矿掉落,而且使侧板稳定性变差,极易引发安全生产事故。针对烧结机台车侧板安全检测问题,设计台车侧板裂缝检测方案,实时获取侧板图像数据,对比U-Net、金字塔场景解析网络(pyramid scene parsing network,PSPNet)、DeepLab V3+三种不同的语义分割网络,试验结果表明,U-Net网络性能较优,对目标特征提取较好,故采用U-Net对侧板边界及裂缝进行检测。在此基础上,根据裂缝类型、长度及数量设定侧板裂缝安全预警判断规则,构建台车侧板裂缝安全预警检测算法,通过试验验证算法总误检率为0.91%。结果表明,该算法为烧结机台车侧板裂缝安全预警提供技术方案,能降低生产事故,对烧结生产安全具有重要意义。The side plate of sinter trolley is easy to produce cracks in the production process,which will not only cause the sinter to fall,but also make the stability of side plate worse.It is easy to cause safety production accidents.Aiming at the safety detection problem of side plate,a crack detection scheme of side plate was designed and the real-time image data of side plate was obtained.Compared with the three different semantic segmentation networks of U-Net,PSPNet,DeepLab V3+,the experimental results prove that U-Net network have better performance and good extraction of target features.Therefore,U-Net was used to detect the side plate boundary and cracks.On this basis,the safety early-warning rules of side plate cracks were set according to the type,length and number of cracks.After that,the safety early-warning detection algorithm for trolley side plate cracks was constructed and total false detection rate of algorithm is 0.91%.The results show that the algorithm can provide technical solutions of safety early-warning for sinter trolley side plate cracks.It can reduce production accidents effectively and has a great significance for sintering production safety.

关 键 词:图像处理 烧结机台车 语义分割 裂缝识别 

分 类 号:TF046.4[冶金工程—冶金物理化学]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象