基于孪生支持向量回归机的转炉炼钢终点预测
|
摘要
为了提高转炉炼钢的终点命中率,建立了一种新的转炉终点预测模型,实现了对转炉终点碳质量分数和温度的准确预测。模型采用K最近邻孪生支持向量机(KNNWTSVR)算法,将权重矩阵引入到目标函数中,并利用鲸群优化算法进行求解,提高了传统算法的性能;然后基于某炼钢厂260t转炉的实际生产数据,建立了转炉炼钢终点预测模型。结果表明,预测模型的终点碳质量分数(误差±0.005%)和温度(误差±15℃)的终点单命中率分别为94%和88%,双命中率达到84%。与其他两种现有的建模方法相比,本模型取得了最优的预测效果。该方法满足转炉炼钢实际生产的需求,也可适用于钢铁冶金其他领域的数学建模。
In order to improve the end-point hit rate of basic oxygen furnace(BOF)steelmaking,a new prediction model of BOF end-point is established to achieve the accurate prediction of the carbon content and temperature at the end of the converter.K-nearest neighbor weighted based twin support vector regression(KNNWTSVR)is adopted for the model.A KNN weighted matrix is introduced to the objective functions,and the whale optimization algorithm is used to solve the objective functions to improve the performance of the algorithm.Then,based on the datasets of a 260 tBOF,the prediction model for converter steelmaking end-point is established.The experimental results show that the predicted hit rates of the end-point carbon content(error±0.005%)and temperature(error±15℃)are 94%and 88%,respectively,and the double hit rate achieves 84%.Compared with the other two existing modeling methods,the proposed model obtains the best prediction effect.Therefore,it meets the requirements of the real production of converter steelmaking,and it is also suitable for mathematical modeling in other fields of metallurgy.
In order to improve the end-point hit rate of basic oxygen furnace(BOF)steelmaking,a new prediction model of BOF end-point is established to achieve the accurate prediction of the carbon content and temperature at the end of the converter.K-nearest neighbor weighted based twin support vector regression(KNNWTSVR)is adopted for the model.A KNN weighted matrix is introduced to the objective functions,and the whale optimization algorithm is used to solve the objective functions to improve the performance of the algorithm.Then,based on the datasets of a 260 tBOF,the prediction model for converter steelmaking end-point is established.The experimental results show that the predicted hit rates of the end-point carbon content(error±0.005%)and temperature(error±15℃)are 94%and 88%,respectively,and the double hit rate achieves 84%.Compared with the other two existing modeling methods,the proposed model obtains the best prediction effect.Therefore,it meets the requirements of the real production of converter steelmaking,and it is also suitable for mathematical modeling in other fields of metallurgy.
引文
[1]高绪东.BP神经网络在高炉铁水硅预报中的应用[J].中国冶金,2014,24(6):24.
[2]崔桂梅,李静,张勇,等.基于T-S模糊神经网络模型的高炉铁水温度预测建模[J].钢铁,2013,48(11):11.
[3]任彦军,王家伟,张晓兵,等.基于LM算法BP神经网络的高炉-转炉界面铁水温度预报模型[J].钢铁,2012,47(9):40.
[4]赵路朋,吴铿,朱利,等.基于BP神经网络的烧结矿性能预报模型[J].钢铁,2017,52(9):11.
[5]丁容,刘浏.转炉炼钢过程人工智能静态控制模型[J].钢铁,1997,32(1):22.
[6]李小环.转炉终点残锰量控制及优化[J].中国冶金,2016,26(5):57.
[7]Frattini Fileti A M,Pacianotto T A,Pitasse Cunha A.Neural modeling helps the BOS process to achieve aimed end-point conditions in liquid steel[J].Eng Appl Artif Intell,2006,19(1):9.
[8]Jayadeva,Khemchandani R,Chandra S.Twin support vector machines for pattern classification[J].IEEE Trans Pattern Anal Machine Intell,2007,29(5):905.
[9]PENG X.TSVR:An efficient twin support vector machine for regression[J].Neural Netw,2010,23(3):365.
[10]XU Y,WANG L.K-nearest neighbor-based weighted twin support vector regression[J].Appl Intell,2014,41(1):299.
[11]XU Y,LI X,PAN X,et al.Asymmetric v-twin support vector regression[J].Neural Comp Appl,2017(2):1.
[12]Mirjalili S,Lewis A.The whale optimization algorithm[J].Advances in Engineering Software,2016,95:51.
[13]贾俊平,何晓群,金勇.统计学[M].4版.北京:中国人民大学出版社,2009.
[2]崔桂梅,李静,张勇,等.基于T-S模糊神经网络模型的高炉铁水温度预测建模[J].钢铁,2013,48(11):11.
[3]任彦军,王家伟,张晓兵,等.基于LM算法BP神经网络的高炉-转炉界面铁水温度预报模型[J].钢铁,2012,47(9):40.
[4]赵路朋,吴铿,朱利,等.基于BP神经网络的烧结矿性能预报模型[J].钢铁,2017,52(9):11.
[5]丁容,刘浏.转炉炼钢过程人工智能静态控制模型[J].钢铁,1997,32(1):22.
[6]李小环.转炉终点残锰量控制及优化[J].中国冶金,2016,26(5):57.
[7]Frattini Fileti A M,Pacianotto T A,Pitasse Cunha A.Neural modeling helps the BOS process to achieve aimed end-point conditions in liquid steel[J].Eng Appl Artif Intell,2006,19(1):9.
[8]Jayadeva,Khemchandani R,Chandra S.Twin support vector machines for pattern classification[J].IEEE Trans Pattern Anal Machine Intell,2007,29(5):905.
[9]PENG X.TSVR:An efficient twin support vector machine for regression[J].Neural Netw,2010,23(3):365.
[10]XU Y,WANG L.K-nearest neighbor-based weighted twin support vector regression[J].Appl Intell,2014,41(1):299.
[11]XU Y,LI X,PAN X,et al.Asymmetric v-twin support vector regression[J].Neural Comp Appl,2017(2):1.
[12]Mirjalili S,Lewis A.The whale optimization algorithm[J].Advances in Engineering Software,2016,95:51.
[13]贾俊平,何晓群,金勇.统计学[M].4版.北京:中国人民大学出版社,2009.