引用本文:韩敏,赵耀,杨溪林,林东.基于鲁棒相关向量机的转炉炼钢终点预报模型[J].控制理论与应用,2011,28(3):343~350.[点击复制]
HAN Min,ZHAO Yao,YANG Xi-lin,LIN Dong.Endpoint prediction model of basic oxygen furnace steelmaking based on robust relevance-vector-machines[J].Control Theory and Technology,2011,28(3):343~350.[点击复制]
基于鲁棒相关向量机的转炉炼钢终点预报模型
Endpoint prediction model of basic oxygen furnace steelmaking based on robust relevance-vector-machines
摘要点击 1807  全文点击 1520  投稿时间：2009-08-20  修订日期：2010-04-26
查看全文  查看/发表评论  下载PDF阅读器
DOI编号  10.7641/j.issn.1000-8152.2011.3.CCTA091068
  2011,28(3):343-350
中文关键词  转炉炼钢  终点预报  相关向量机  噪声方差系数
英文关键词  BOF steelmaking  endpoint prediction  relevance-vector-machines  coefficient of noise variance
基金项目  国家高技术研究发展计划“863”计划资助项目(2007AA04Z158); 国家自然科学基金资助项目(60674073).
作者单位E-mail
韩敏* 大连理工大学 电子信息与电气工程学部 minhan@dlut.edu.cn 
赵耀 大连理工大学 电子信息与电气工程学部  
杨溪林 北京金自天正智能控制股份有限公司  
林东 本钢板材股份有限公司炼钢厂  
中文摘要
      针对传统相关向量机在训练过程中易受异常点影响的问题, 提出了一种鲁棒相关向量机模型, 并将其应用于转炉炼钢终点碳含量和温度的预报. 通过为每一个训练样本设定独立的噪声方差系数, 并使其在训练过程中随模型预测误差的增大而逐渐减小来降低异常点的影响, 同时依据贝叶斯证据框架给出了模型超参数的迭代计算公式, 进行参数的优化. 使用标准测试数据和转炉炼钢实际生产数据进行仿真, 结果表明本文模型具有较好的预报精度和鲁棒性.
英文摘要
      To deal with the problem that the classical relevance vector machine is sensitive to outliers, we present a novel robust relevance vector machine. This machine is applied to predict the endpoint carbon content and temperature of the basic-oxygen-furnace(BOF) steelmaking. Each training sample is assumed to have its individual coefficient of noise variance. With the increase of the prediction error during training procedure, the coefficients of outliers gradually decrease, reducing the impact of outliers. In addition, the iterative formulas for the optimization of hyper-parameters are derived in the Bayesian evidence framework. Simulation results of benchmark test data and the BOF steelmaking data show that the proposed mode achieves high prediction accuracy and good robustness.