基于多源图像解析的钢铁冷轧退火炉元器件多目标测温预警  

MULTI-OBJECTIVE TEMPERATURE MONITORING FOR STEEL COLD ROLLING ANNEALING FURNACE COMPONENTS BASED ON MULTI-SOURCE IMAGE ANALYSIS

作　　者：薛佳朋 王钊哲 丁泽浩 亢克松 李子晗 Xue Jiapeng;Wang Zhaozhe;Ding Zehao;Kang Kesong;Li Zihan(School of Computer Science and Technology,Shandong University,Qingdao 266000,Shandong;Xiongan WeShyper Intelligent Technology Co.,Ltd.,Baoding 071000,Hebei;Intelligent Manufacturing Division of HBIS Group Digital Technology Co.,Ltd.,Tangshan 063000,Hebei)

机构地区：[1]山东大学计算机学院,山东青岛266600 [2]雄安威赛博智能科技有限公司,河北保定071000 [3]河钢数字技术股份有限公司智能制造事业部,河北唐山063000

出　　处：《河北冶金》2024年第11期25-29,43,共6页Hebei Metallurgy

摘　　要：在经济高质量发展以及国家积极推动传统产业转型升级的背景下,钢铁行业面临着产能过剩、生产流程复杂、质量管控难等问题,需要融入新一代信息技术来推动钢铁产业的转型升级。在钢铁企业的生产过程中,设备的平稳运行对企业的生产效率、产品质量、安全保障及经济效益至关重要。针对钢铁企业生产设备种类繁多、高温环境导致设备损耗高、现场维保危险性大等特点,提出了基于红外热成像和计算机视觉技术的冷轧退火炉元器件温度非接触式自动监测预警系统。重点研究了基于计算机视觉技术的多目标检测和图像语义分割算法与红外热成像相结合对退火炉元器件实现多元图像解析,获取元器件目标的实时图像与实测温度的过程。该系统可以实现对炼钢设备温度的实时监测和管理,提高钢铁企业的安全生产水平,进一步实现钢铁企业设备的智能化监测和管理,有利于减少安全风险、提高生产效率、保证产品质量和经济效益,推动钢铁行业的高质量发展。In the context of high-quality economic development and the country's active promotion of the transformation and upgrading of traditional industries,the steel industry is faced with overcapacity,complex production processes,difficult quality control and other problems,it is necessary to integrate a new generation of information technology to promote the transformation and upgrading of the steel industry.In the production process of iron and steel enterprises,the smooth operation of equipment is crucial to the production efficiency,product quality,safety and economic benefits of enterprises.A non-contact automatic temperature monitoring and early warning system for cold rolling annealing furnace components based on infrared thermal imaging and computer vision technology was proposed in view of the various types of production equipment in iron and steel enterprises,high equipment loss caused by high temperature environment and high risk of on-site maintenance.This paper focuses on the process of multi-target detection and image semantic segmentation based on computer vision technology combined with infrared thermal imaging to realize multi-image analysis of annealing furnace components,and obtain real-time image and measured temperature of components.The system can realize the real-time monitoring and management of the temperature of steelmaking equipment,improve the safety level of iron and steel enterprises,and further realize the intelligent monitoring and management of iron and steel enterprises equipment,which is conducive to reducing safety risks,improving production efficiency,ensuring product quality and economic benefits,and promoting the high-quality development of the steel industry.

关 键 词：冷轧 退火炉 红外热成像 视觉技术 目标检测 语义分割算法 

分 类 号：TG155.1[金属学及工艺—热处理]