| 基于组合训练方法的RBFNN转炉炼钢静态模型 |
| |
| 引用本文： | 朱亚萍,王文龙,徐峥,徐生林. 基于组合训练方法的RBFNN转炉炼钢静态模型[J]. 杭州电子科技大学学报, 2011, 31(3): 62-65 |
| |
| 作者姓名： | 朱亚萍 王文龙 徐峥 徐生林 |
| |
| 作者单位： | 1. 杭州电子科技大学,自动化学院,浙江杭州,310018
2. 杭州市七格污水处理厂工程建设指挥部,浙江杭州,310019 |
| |
| 基金项目： | 省公益技术应用研究资助项目(C31016) |
| |
| 摘 要： | 为了提高转炉炼钢终点碳含量和温度的预报命中率,该文采用径向基神经网络建立转炉炼钢静态模型.量子微粒群优化算法具体较好的全局搜索能力,而梯度下降法有较好的局部搜索能力,为了能够发挥这两种算法的优势,该文提出了一种组合训练方法,用来训练径向基神经网络.并通过对某炼钢厂的历史数据进行仿真实验,比较组合训练方法与非组合训练方法...
|
| 关 键 词： | 径向基神经网络 转炉炼钢 梯度下降法 量子微粒群优化算法 |
A Static Model of Converter Steelmaking Using RBFNN Based on Combined Methods |
| |
| ZHU Ya-ping,WANG Wen-long,XU Zheng,XU Sheng-lin. A Static Model of Converter Steelmaking Using RBFNN Based on Combined Methods[J]. Journal of Hangzhou Dianzi University, 2011, 31(3): 62-65 |
| |
| Authors: | ZHU Ya-ping WANG Wen-long XU Zheng XU Sheng-lin |
| |
| Affiliation: | ZHU Ya-ping1,WANG Wen-long1,XU Zheng2,XU Sheng-lin1(1.School of Automation,Hangzhou Dianzi University,Hangzhou Zhejiang 310018,China,2.Construction Headquarters of Sewage Treatment Plant in Hangzhou Qige,Hangzhou Zhejiang 310019,China) |
| |
| Abstract: | In order to increase the hit rate of the BOF endpoint,a static model of converter steelmaking was established using radial basis function neural network.Quantum-behaved particle swarm optimization has better global search capability,but gradient descent method has better local search capability.To be able to play the advantages of the two algorithms,a combination of training methods was proposed,to train the radial basis function neural network.Stimulation was done based on the historical data of a steel pl... |
| |
| Keywords: | radial basis function neural network converter steelmaking gradient descent method quantum-behaved particle swarm optimization |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
