| 模糊贝叶斯网络应用于预测高炉铁水含硅量变化趋势 |
| |
| 引用本文： | 龚淑华,刘祥官. 模糊贝叶斯网络应用于预测高炉铁水含硅量变化趋势[J]. 冶金自动化, 2005, 29(5): 30-32,42 |
| |
| 作者姓名： | 龚淑华 刘祥官 |
| |
| 作者单位： | 浙江大学,数学系,浙江,杭州,310027 |
| |
| 基金项目： | 国家级科技成果重点推广计划项目（99040422A） |
| |
| 摘 要： | 贝叶斯网络在高炉铁水含硅量预测中已取得较好效果.本文的进一步改进是利用模糊逻辑方法能很好地将数据分成离散模糊集的优势,对模型参数进行有效的模糊分类,以此作为贝叶斯网络的输入,进行混合建模.对山东莱钢1号高炉智能控制专家系统在线采集数据进行计算证明,对一般高炉混合模型可提高预测命中率到90%.
|
| 关 键 词： | 贝叶斯网络 模糊逻辑 预测 铁水含硅量 |
| 文章编号： | 1000-7059（2005）05-0030-03 |
| 收稿时间： | 2005-03-10 |
| 修稿时间： | 2005-03-102005-05-10 |
Application of fuzzy Bayesian network to trend prediction of silicon content in molten iron |
| |
| GONG Shu-hua,LIU Xiang-guan. Application of fuzzy Bayesian network to trend prediction of silicon content in molten iron[J]. Metallurgical Industry Automation, 2005, 29(5): 30-32,42 |
| |
| Authors: | GONG Shu-hua LIU Xiang-guan |
| |
| Affiliation: | Mathematical Department of Zhejiang University, Hangzhou 310027, China |
| |
| Abstract: | Quite good result has been obtained in the prediction of silicon content in molten iron because of application of Bayesian network. Fuzzy logic is an effective method to divide continuous data into discrete fuzzy sets. Fuzzy clustering of model parameters is made by use of fuzzy logic method. Then, data after clustering is taken as input of Bayesian network to form a hybrid mathematical model. Through calculation of realtime data collected from intelligent control expert system in No 1 BF of Laiwu Iron and Steel Group Co, it is proven that the hybrid model can improve hit rate to 90 % for common BF. |
| |
| Keywords: | Bayesian network fuzzy logic prediction silicon content in hot metal |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
