首页> 外国专利> Apparatus for learning tissue composition of sintered ore, method for learning tissue composition and program for learning organization composition, and apparatus for changing sinter manufacturing condition, manufacturing condition changing method and manufacturing condition change program

Apparatus for learning tissue composition of sintered ore, method for learning tissue composition and program for learning organization composition, and apparatus for changing sinter manufacturing condition, manufacturing condition changing method and manufacturing condition change program

机译:用于学习烧结矿石组织组织组织的装置,用于学习组织组合物的方法和学习组织组成的程序,以及用于改变烧结制造条件的装置,制造条件改变方法和制造条件变化计划

摘要

Problem to be solved: to provide an organization configuration learning apparatus capable of estimating the tissue composition of the sintered ore efficiently and precisely with respect to the tissue composition of the sintered ore.The tissue composition learning apparatus 1 of the sintered oreAn image data acquisition section 21 for acquiring a plurality of image data showing a plurality of sectional images obtained by imaging a cross-section of a sintered ore, andFor each of the multiple regions of the sectional imageHematiteMagnetiteA calcium ferrite and silicate slag and a data generating section 23 for generating a plurality of learning data in association with a poreA learning processing unit 24 which learns the relationship between a plurality of cross sectional images and the composition of sinter by using learning data is learned.Based on the relationship between multiple cross-sectional images learned and the composition of sinterThe learning model output section 25 has a learning model output section 25 for outputting a learning model for estimating the composition of the sintered ore from a cross-sectional image obtained by imaging a cross section of the sintered ore.Diagram
机译:要解决的问题:提供一种组织配置学习装置,其能够有效且精确地相对于烧结矿体的组织组成估计烧结矿石的组织组成。烧结口头图像数据采集部分的组织组成学习装置1 21,用于获取多个图像数据,示出通过成像烧结矿石的横截面而获得的多个截面图像,并且对于截面剖视图的多个区域中的每一个用于产生用于产生a的数据产生部分23与孔隙学习处理单元24相关联的多个学习数据,其学习通过使用学习数据的多个横截面图像和烧结组合物之间的关系。基于多个横截面图像之间的关系和组合物Sinterthe学习模型输出部分25有一个lea越型模型输出部分25用于输出用于从通过成像烧结矿石的横截面而获得的横截面图像来估计烧结矿石的组成的学习模型

著录项

  • 公开/公告号JP2021165429A

    专利类型

  • 公开/公告日2021-10-14

    原文格式PDF

  • 申请/专利权人 日本製鉄株式会社;

    申请/专利号JP20200069767

  • 发明设计人 水谷 守利;

    申请日2020-04-08

  • 分类号C22B1/16;G06T7;

  • 国家 JP

  • 入库时间 2022-08-24 21:42:17

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号