期刊文献+

深度学习算法在钢铁质量检测中的应用

Application of depth learning algorithm in steel quality inspection
下载PDF
导出
摘要 我国钢铁产业正经历产业结构优化升级、钢铁产品由量到质的革命性阶段。高效率、高精度且具有普适性质的深度学习算法对复杂钢铁产品质量检查具有重要意义。针对目前钢铁行业质量检查中不可逆损耗、抽样效率低、人工成本高和检测可靠性差等关键技术挑战,文章创新性地通过整合组批算法、性能预测模型,提升了网络运算速度、钢铁产品检测效率,进而大大地降低了检测成本。基于日钢营口中板有限公司中厚板改造项目的实践数据验证,文章提出的算法能够满足于生产实际,所带来的经济效益远远高于传统的计算模型和人工检测方法,对复杂环境下的系统钢铁质量检测具有重要的现实意义。 Our national iron and steel industry is going through a revolutionary stage in which the industrial structure is adjusted and upgraded while iron and steel products are transformed from quantity to quality.High-efficiency,high-precision,and universal deep learning algorithms are of great significance for the complicated procedure of steel product inspections.This work integrates batching algorithms and performance prediction models to improve network computing speed and steel products,aiming to solve the key technical challenges in the current quality inspection of the steel industry,such as irreversible loss,low sampling efficiency,high labor cost,and poor detection reliability.Our model showed promising capabilities to increase detection efficiency and reduce final economic costs.Based on the experimental data generated by a heavy plate reconstruction project in Rizhao Steel Yingkou Medium Plate Co.,Ltd.,we demonstrate that the algorithm proposed in this work can satisfy the actual production.The resulted economic benefits are far higher than that generated using the traditional calculation model and manual detection method,which shows great potential in complicated product detection.
作者 耿朝雷 艾云霄 Geng Chaolei;Ai Yunxiao(Beijing Jingcheng Dingyu Management System Co.,Ltd.,Beijing 100176,China)
出处 《无线互联科技》 2022年第18期96-99,共4页 Wireless Internet Technology
关键词 检验批 计划组批 动态组批 深度学习 性能预测 CNN-LSTM模型 inspection lot plan group approval dynamic batching deep learning performance prediction CNN-LSTM model
  • 相关文献

参考文献5

二级参考文献60

  • 1刘晓冰,李松,蒙秋男,郭燕锋.面向MES的特钢企业标准化管理系统的研究与开发[J].冶金标准化与质量,2004,42(6):28-31. 被引量:1
  • 2《建设工程监理范围和规模标准规定》、《建设工程监理规范》知识竞赛结果[J]建设监理,2001(03).
  • 3GB 50300-2001.建筑工程施工质量验收统一标准[S],2001.
  • 4GB 50319-2000.建设工程监理规范[S],2000.
  • 5Rabiner L.Atutprial on hidden Markov models and selected applications in seeech recognition[J].Proceedings of the IEEE,1989,77(2):257-286.
  • 6Bar-Shalon Y,Fortmann TE.Tracking and Data Association[M].New York:Academic Press,1998.
  • 7Comaniciu D,Ramesh V,Meer P.Real-time tracking of non-rigid objects using mean-shift[C]//Proceedings of the IEEE Conference Computer Vision and Patter Recogition,2000:142-149.
  • 8Isard M,Blake A.Condensation-conditional density propaga-tion for visual tracking[J].International Journal of Computer Vision,1998,29(1):5-28.
  • 9WU Yi,LIM Jongwoo,YANG M H.Online object tracking:A benchmark[C].IEEE Conference on Computer Vision and Pattern Recognition,2013:1354-1362.
  • 10ROSS D,LIM J,LIN R,et al.Incremental learning forrobust visual tracking[J].International Journal of Compute Vision,2008,77(1):125-141.

共引文献5

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部