泛研网

公众版 查看说明
全球科研项目数据库2026版

基于计算机视觉的连铸漏钢空间—时序特征可视化及其智能预测集成方法

项目来源

国家自然科学基金(NSFC)

项目主持人

刘宇

项目受资助机构

东北电力大学

立项年度

2017

立项时间

未公开

项目编号

51704073

项目级别

国家级

研究期限

未知 / 未知

受资助金额

20.00万元

学科

工程与材料科学-矿业与冶金工程-钢铁冶金

学科代码

E-E04-E0411

基金类别

青年科学基金项目

关键词

预测 ; 结晶器 ; 漏钢 ; 计算机视觉 ; 连铸 ; continuous casting ; mould ; breakout ; prediction ; computer vision

参与者

狄驰；王俊尧；王福旺；韩璐潞；金娜

参与机构

东北电力大学

项目标书摘要：漏钢是连铸生产的灾难性事故，及时、准确预报漏钢是稳定和保障连铸顺行的重中之重。目前频繁误报、漏报等问题普遍存在，一维温度曲线预报方法已无法满足持续、高效的连铸发展需求。因此，本项目重点围绕连铸漏钢空间—时序特征可视化和预测集成方法进行研究。首先，在细致考察保护渣、钢水成分等因素对漏钢影响和作用的基础上，借助帧间差分算法，实现结晶器温度热成像，运用区域分割、游程标记等计算机视觉算法，快速、准确提取漏钢空间—时序可视化特征。其次，基于积累的漏钢样本，统计归纳漏钢的共性可视化特征，揭示漏钢空间传播和动态演化规律。最后，采用决策树对异常区域进行预判断，剔除部分伪黏结区域，建立随机森林、支持向量机漏钢预测模型，通过训练和测试，确立准确、高效的漏钢预测方法。研究结果有望在预防、预测漏钢事故上取得新进展，开发基于视觉—空间/时序—人工智能的漏钢预测集成新方法，提升连铸过程异常检测的可视化、智能化水平。

Application Abstract: Breakout is a disastrous accident during continuous casting.The accuracy and speediness of breakout prediction are important to the smooth and steady continuous casting production.At present,because of the frequent false and missed alarms,the prediction method of one-dimensional temperature cannot satisfy the needs of quick and sustainable development of continuous casting.Therefore this project focuses on breakout space-time characteristics visualization and integrated prediction method during continuous casting.Firstly,a mould thermograph is developed by virtue of frame difference algorithm based on careful investigation on the influence factors of breakouts,such as mould powder and molten steel composition.Then visual space-time characteristics are extracted quickly and accurately by computer vision algorithms,such as region segmentation,run labeling.Secondly,the common visual characteristics are epurated based on the statistics of the collected breakout samples.The space propagation and dynamic evolution rule of breakouts will be revealed.Finely,a decision tree model will be adopted to prejudge the abnormal regions in order to remove some false sticker regions.Then random forest and support vector machine models will be built up,trained and tested in order to establish an accurate and efficient prediction method.The results are helpful to obtain new progress on prevention and prediction of breakout.An integrated breakout prediction method is developed by computer vision,space-time characteristics,artificial intelligence.It also improves the visualization and intelligence of abnormal detection during continuous casting.

项目受资助省

吉林省

项目结题报告(全文)

漏钢是连铸板坯生产的重大安全事故，频繁的误报和时有的漏报严重影响了连铸生产顺行。本项目基于连铸结晶器在线监控系统和现场实验研究，从钢种、铸坯尺寸、保护渣、拉速、液位等方面，着重分析了黏结漏钢的主要影响因素，在此基础之上，利用计算机视觉检测技术，采用游程标记、区域分割等图像处理算法，对黏结漏钢异常区域进行了快速标记，并提取热区域和冷区域特征，之后，基于现场的黏结漏钢实例，统计归纳了三种形态的黏结漏钢时空特征，揭示了漏钢空间传播和动态演化规律，最后，采用决策树方法，对图像中的感兴趣区域特征进行快速提取，以感兴趣区域的热区域和冷区域特征为输入，建立支持向量机、随机森林智能化模型。研究结果表明：低合金钢比低碳钢更容易发生黏结漏钢，黏结漏钢实例中液位波动超过20mm的全部为低合金钢，结晶器内存在典型黏结漏钢、单侧传播黏结漏钢和大跨度黏结漏钢三种形态，其中，单侧传播黏结漏钢具有最大的升温速率和最低的降温速率，黏结漏钢具有典型的热区域和冷区域，黏结漏钢的热区域和冷区域面积、温度速率和位置特征的组合更适合作为判断真伪黏结漏钢的依据，以真伪黏结漏钢空间—时序特征为输入，建立了支持向量机和随机森林模型，漏钢预测准确率为93.8%。截止到2020年12月31日，本项目研究已发表学术论文13篇，其中,SCI检索论文11篇,EI检索论文2篇，出版学术专著1部，授权发明专利1项，审查中发明专利1项，授权软件著作权1项。项目研究对深刻认识结晶器内黏结漏钢形成和演化规律，提高连铸漏钢预报准确率，提升连铸结晶器在线监控的可视化、智能化水平具有重要的科学意义，在连铸工艺参数优化、漏钢预报等方面具有较好的应用前景。

  • 排序方式：
  • 1
  • /

  • 1.Intelligent breakout prediction method based on support vector machine

    • 关键词：
    • Accidents;Support vector machines;Continuous casting;Forecasting;Accuracy rate;Alarm rate;Breakout predictions;False alarms;Spatial characteristics;Support vector machine models;Temperature data;Temperature signal
    • Tian, Yuanpeng;Liu, Yu
    • 《2020 International Conference on Advanced Materials and Intelligent Manufacturing and Advanced Steel for Automotive Seminar, ICAMIM 2020》
    • 2020年
    • August 21, 2020 - August 23, 2020
    • Guilin, China
    • 会议

    Breakout is a serious accident during continuous casting. In order to reduce breakout accident, a breakout prediction method was proposed based on support vector machine model. The typical characteristics of continuous casting breakout were reconstructed by thermocouple temperature signals. The temperature data of three adjacent thermocouples in the first row were added and then connected with the temperature data of the middle second row, which was used to capture the typical spatial characteristics of breakout. The 80% of true and false breakout samples were used to train the support vector machine model and 20% samples were used to test the model. The results show that this method can predict all breakouts, which has a true-sticker alarm rate of 100%. There is one false alarm and the accuracy rate is 95.8%.
    © 2020 Institute of Physics Publishing. All rights reserved.

    ...

  • 2.Research on microstructure and wear property of 45 steel quenched with different laser power

    • 关键词：
    • Wear of materials;Quenching;Carbon dioxide lasers;Hardness;Friction and wear properties;Hardness increase;Laser quenching;Laser scanning speed;Layer thickness;Maximum hardness;Microstructure and properties;Wear properties
    • Yu, Liu;Tianhao, Xu;Ying, Liu;Haicheng, Zhang;Xingxing, Li;Guangbin, Chen
    • 《9th International Conference on Advanced Materials and Engineering Materials, ICAMEM 2020》
    • 2020年
    • July 3, 2020 - July 5, 2020
    • Bangkok, Thailand
    • 会议

    The surface of 45 steel is quenched by CO2 laser with scanning speed 1000 mm/min and different laser power 1000W, 1200W, 1400W, 1600W and 1800W. Experiments are carried out to analyze microstructure, friction and wear properties of quenched 45 steel. The results show that the quenching layer thickness increases gradually with the increase of laser power, and the maximum value of quenching layer hardness increases first and then decreases. When the laser power is 1600W, the maximum hardness value is 883HV0.5. But when the laser power is 1800W, the hardness of quenching layer becomes to decrease. The reason is the surface of 45 steel becomes to melt. The wear volume increases first and then decreases too. When laser power is 1600W, the minimum wear volume is 0.08mm3, which is 6.4% to the wear volume of 45 steel matrix without laser quenching. Therefore, better microstructure and properties of 45 steel can be obtained when laser scanning speed is 1000mm/min and laser power is 1600W.
    © 2020 Trans Tech Publications Ltd, Switzerland.

    ...

  • 3.Research on microstructure and wear property of 45 steel quenched with different laser power

    • 关键词：
    • Wear of materials;Quenching;Carbon dioxide lasers;Hardness;Friction and wear properties;Hardness increase;Laser quenching;Laser scanning speed;Layer thickness;Maximum hardness;Microstructure and properties;Wear properties
    • Yu, Liu;Tianhao, Xu;Ying, Liu;Haicheng, Zhang;Xingxing, Li;Guangbin, Chen
    • 《9th International Conference on Advanced Materials and Engineering Materials, ICAMEM 2020》
    • 2020年
    • July 3, 2020 - July 5, 2020
    • Bangkok, Thailand
    • 会议

    The surface of 45 steel is quenched by CO2 laser with scanning speed 1000 mm/min and different laser power 1000W, 1200W, 1400W, 1600W and 1800W. Experiments are carried out to analyze microstructure, friction and wear properties of quenched 45 steel. The results show that the quenching layer thickness increases gradually with the increase of laser power, and the maximum value of quenching layer hardness increases first and then decreases. When the laser power is 1600W, the maximum hardness value is 883HV0.5. But when the laser power is 1800W, the hardness of quenching layer becomes to decrease. The reason is the surface of 45 steel becomes to melt. The wear volume increases first and then decreases too. When laser power is 1600W, the minimum wear volume is 0.08mm3, which is 6.4% to the wear volume of 45 steel matrix without laser quenching. Therefore, better microstructure and properties of 45 steel can be obtained when laser scanning speed is 1000mm/min and laser power is 1600W.
    © 2020 Trans Tech Publications Ltd, Switzerland.

    ...

  • 4.Microstructure and properties of Cr12MoV die steel by laser quenching with different power

    • 关键词：
    • Ternary alloys;Carbon dioxide lasers;Quenching;Chromium steel;Microstructure;Wear of materials;Molybdenum alloys;Hardness;Vanadium alloys;Cold work die;Friction and wear properties;Laser quenching;Laser scanning speed;Maximum hardness;Microstructure and properties;Original sample;Scanning speed
    • Yu, Liu;Yuanpeng, Tian;Hui, Zhang;Haijing, Zhang;Yang, Li
    • 《2019 5th International Conference on Applied Materials and Manufacturing Technology, ICAMMT 2019》
    • 2019年
    • June 21, 2019 - June 23, 2019
    • Singapore, Singapore
    • 会议

    In this paper, laser quenching on the surface of Cr12MoV cold-work die steel was carried out by CO2 laser. The laser scanning speed was 800mm/min, and the laser powers were 1200W, 1400W, 1600W and 1800W respectively. Thickness, microstructure, hardness, XRD, friction and wear properties of the quenched layer were investigated after laser quenching with different power. The results show that the thickness and martensite content in quenched layer increase gradually with the increase of laser power. The hardness of quenched layers have an increase firstly and a later decrease along with laser power. When the laser power is 1600W, the maximum hardness is 662.9HV. When the power is 1800W, microfusion appears on the surface, and the hardness of the quenching layer decreases. The wear volume of the sample first rises and then falls along with laser power. The minimum wear volume is 0.09mm3 at 1600W, and the wear volume is reduced by 92% of the original sample. Therefore, scanning speed of 800mm/min and laser power of 1600W can get the best microstructure and properties.
    © Published under licence by IOP Publishing Ltd.

    ...
  • 排序方式：
  • 1
  • /