基于改进BP神经网络的连铸漏钢预报

中国机械工程 ›› 2012, Vol. 23 ›› Issue (2): 204-207.

基于改进BP神经网络的连铸漏钢预报

张本国1;李强1,2;王葛1;张水仙1

1.燕山大学,秦皇岛,066004

2.河北科技大学,石家庄,050018

出版日期:2012-01-25 发布日期:2012-02-14
基金资助:
河北省科学技术研究与发展计划资助项目（07212119D）
Hebei Provincial S&T Research and Development Program of China（No. 07212119D）

Breakout Prediction Based on Improved BP Neural Network in Continuous Casting Process

Zhang Benguo1;Li Qiang1,2;Wang Ge1;Zhang Shuixian1

1.Yanshan University,Qinhuangdao,Hebei,066004

2.Hebei University of Science and Technology,Shijiazhuang,050018

Online:2012-01-25 Published:2012-02-14
Supported by:
Hebei Provincial S&T Research and Development Program of China（No. 07212119D）

摘要/Abstract

摘要：

针对传统BP神经网络在训练过程中存在收敛速度慢的缺陷,将LM(levenberg marquardt)算法引入到BP神经网络的训练过程,建立了LM-BP神经网络模型,并将其应用于连铸过程中的漏钢预报系统。结合某钢厂连铸现场历史数据对系统进行了测试,测试结果以96.15%的预报率及100%的报出率,验证了基于LM算法的BP神经网络连铸漏钢预报方案的可行性和有效性。



关键词:

连铸, 漏钢预报, LM算法, BP神经网络

Abstract:

LM algorithm was introduced to the training process of

a BP neural network and a LM-BP neural network model was established aiming at the

defects of slow convergence in the training process of the traditional BP

neural network.The LM-BP neural network model was applied to the breakout prediction

in the continuous casting processes,and it was tested with the historical data collected

from a steel mill.The feasibility and the validity of the model are verified by the results with

the accuracy rate of 96.15% and the prediction rate of 100%.

Key words:

continuous casting, breakout prediction, LM(Levenberg Marquardt) algorithm, BP neural network

中图分类号:

TF777.1

张本国1, 李强1, 2, 王葛1, 张水仙1.

基于改进BP神经网络的连铸漏钢预报

[J]. 中国机械工程, 2012, 23(2): 204-207.

ZHANG Ben-Guo-1, LI Jiang-1, 2, WANG Ge-1, ZHANG Shui-Xian-1.

Breakout Prediction Based on Improved BP Neural Network in Continuous Casting Process

[J]. China Mechanical Engineering, 2012, 23(2): 204-207.

导出引用管理器 EndNote|Ris|BibTeX

链接本文: http://www.cmemo.org.cn/CN/

http://www.cmemo.org.cn/CN/Y2012/V23/I2/204

[1]周汉香[1],于学斌[2].连铸漏钢预报技术[J].炼钢,1999,15(4):57-60.
[2]胡志刚,毕学工.连铸结晶器内粘结漏钢的形成机理及对策[J].炼钢,1998,14(6):23-27.
[3]王志政[1],孙健[1],杨晓刚[1].提高板坯连铸漏钢预报系统准确性的措施[J].连铸,2010(3):39-42.
[4]孙立根[1],张家泉[1].基于逻辑判断的板坯漏钢预报系统研究[J].冶金自动化,2009,33(1):16-20.
[5]秦旭[1,2],朱超甫[1],高光河[2],尹延荣[2].板坯漏钢机理及基于热电偶检测的漏钢预报技术研究[J].钢铁研究学报,2011,23(2):7-11.
[6]郭戈,乔俊飞.基于模糊神经网络的一种漏钢预报方法[J].控制理论与应用,1998,15(4):593-598.
[7]王唯一[1],荣亦诚[1],龚幼民[1],孙卓[1],张明[1],耿海[1],傅正权[2].基于模糊聚类的结晶器漏钢动态波形识别及其仿真[J].系统仿真学报,2003,15(4):472-475.
[8]李义宝[1],张学勇[1],马建国[1],汪力君[1].基于BP神经网络的改进算法研究[J].合肥工业大学学报：自然科学版,2005,28(6):668-671. 0
[9]Zhang Li, Luo Jianhua, Yang Suying. Forecasting Box Office Revenue of Movies with BP Neural Net- workEJ~. Expert Systems with Applications, 2009, 36 (3) : 6580-6587.
[10]klsiang S H,Lin Y W. Application of Fuzzy Theo- ry to Predict Deformation Behaviors of Magnesium Alloy Sheets under Hot Extrusion~-J~. Journal of Materials Processing Technology, 2008,201 (1/3) : 138-144.
[11]焦李成.神经网络系统理论[M].西安：西安电子科技大学出版社,1992.
[12]胡洁[1,2],曾祥金[1].一种快速且全局收敛的BP神经网络学习算法[J].系统科学与数学,2010(5):604-610.
[13]高源[1].BP神经网络优化算法研究[J].电脑知识与技术,2009,0(10X):8248-8249.
[14]吕砚山,赵正琦.BP神经网络的优化及应用研究[J].北京化工大学学报：自然科学版,2001,28(1):67-69.
[15]庞清乐[1,2],孙同景[1],杨福刚[1],钟麦英[1].基于粗集理论的归一化方法[J].计算机工程,2007,33(8):36-38.
[16]柳小桐[1].BP神经网络输入层数据归一化研究[J].机械工程与自动化,2010(3):122-123.