基于WKLSC-LWKL相似性度量策略的转炉炼钢终点碳温软测量方法
DOI:
作者:
作者单位:

昆明理工大学

作者简介:

通讯作者:

中图分类号:

TP11

基金项目:

国家自然科学基金资助项目(面向转炉炼钢终点控制的火焰吹炼信息特征提取与熔池碳温连续实时预报模型研究,No.61863018)、云南省科技厅面上项目(基于火焰图像动静态特征提取的转炉炼钢终点吹炼数据实时预报模型研究,No. 202001AT070038)


Soft measurement method of endpoint carbon content and temperature of converter steelmaking based on WKLSC-LWKL similarity measurement strategy
Author:
Affiliation:

Kunming University of Science and Technology

Fund Project:

Project supported by the National Natural Science Foundation of China (Research of flame blowing information features extraction and molten pool carbon temperature continuous re-al-time prediction model for converter steelmaking endpoint control,No.61863018)、the Applied Basic Research Programs of Yunnan Science and Technology Department (Research on real time prediction model of BOF end point blowing data based on dynamic and static feature extraction of flame image,No. 202001AT070038)

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    转炉炼钢终点控制的关键是实现碳温准确预测.针对炉次样本间波动性较大,导致即时学习的样本相似性度量困难而造成预测精度不高的问题,本文提出一种基于改进谱聚类算法构建的相似性度量策略.首先,根据过程变量和关键变量间的耦合关系构造全局加权KL度量准则的谱聚类算法,获得类间方差较大、类内方差较小的聚类子集以消除炉次样本间的波动;其次,根据类簇间差异信息,融入局部加权KL度量准则计算待测样本隶属于各类的后验概率,从而构造出一种适合描述转炉炼钢过程复杂特性的相似性度量策略;最后,采用该度量策略度量出与待测炉次工况特性更加相似的样本子集,建立相关向量机回归模型进行终点碳温预测.实际转炉炼钢生产过程数据仿真结果表明,碳含量在±0.02%的预测误差范围内精度达到89%,温度在±10℃的误差范围内精度达到92%.

    Abstract:

    Accurate prediction of carbon content and temperature is the crucial to the endpoint control of converter steelmaking. For the sample fluctuation is large, it is difficult to measure the similarity of samples in Just-in time Learning, which leads to the problem of low prediction accuracy, this paper proposes a similarity measurement strategy based on an improved spectral clustering algorithm. Firstly, according to the coupling relationship between process variables and dominant variables, a spectral clustering algorithm with global Weighted KL measurement standards was constructed, thus, the clustering subsets with large between clusters variance and small intra-cluster variance were obtained to eliminate the fluctuation among the furnace samples. Secondly, according to the difference information between class clusters, the Local Weighted KL metric criterion was integrated to calculate the posterior probability of the predicted samples belonging to various clusters, then, a similarity measurement strategy suitable for describing the complex characteristics of converter steelmaking process is constructed. Finally, this measurement strategy is used to calculate a subset of samples that are more similar to the properties of the new furnace, and the RVM model is established to predict the end point carbon content, and temperature. The simulation results of actual converter steelmaking process show that prediction accuracy of carbon content within ±0.02 % error range reach 89%, temperature within ±10℃ error range reach 92%.

    参考文献
    相似文献
    引证文献
引用本文
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2021-04-28
  • 最后修改日期:2021-08-02
  • 录用日期:2021-08-09
  • 在线发布日期:
  • 出版日期: