| 基于广义回归神经网络的碳铵塔预测模型 |
| |
| 引用本文： | 郑启富,罗晟,谢艳. 基于广义回归神经网络的碳铵塔预测模型[J]. 化工技术与开发, 2006, 35(4): 37-39 |
| |
| 作者姓名： | 郑启富 罗晟 谢艳 |
| |
| 作者单位： | 浙江工业大学浙西分校化工系,浙江,衢州,324000;巨化集团公司锦纶厂,浙江,衢州,324004 |
| |
| 摘 要： | 采用改进的遗传算法优化广义回归神经网络(GRNN)的平滑参数,并运用GRNN的非线性映射能力,建立了碳铵塔出口碳化度和氨滴度的预测模型.检验结果表明,该模型具有良好的预测性能.
|
| 关 键 词： | 碳铵塔 碳化度 氨滴度 广义回归神经网络 遗传算法 |
| 文章编号： | 1671-9905(2006)04-0037-03 |
| 收稿时间： | 2005-11-04 |
| 修稿时间： | 2005-11-04 |
Prediction Model of Salvolatile Column Based on General Regression Neural Networks |
| |
| ZHENG Qi-fu,LUO Sheng,XIE Yan. Prediction Model of Salvolatile Column Based on General Regression Neural Networks[J]. Technology & Development of Chemical Industry, 2006, 35(4): 37-39 |
| |
| Authors: | ZHENG Qi-fu LUO Sheng XIE Yan |
| |
| Affiliation: | 1. Department of Chemical Engineering, West Branch of Zhejiang University of Technology, Quzhou 324000, China; 2. Caprone Plant of Ju-hua Group Company, Quzhou 324004, China |
| |
| Abstract: | General regression neural networks(GRNN) had a strong ability of simulating non-linear function.It could find the hidden relation between independent variable and dependent variable according to the sample data.The optimization of the smoothing factor was crucial to the performance of GRNN,and it was also the essence and difficulty of GRNN training.A modified genetic algorithm was applied to optimize smoothing factors of GRNN,then model of salvolatile column was built based on GRNN.The model could be applied to predict the carbonization degree and ammonia concentration in the exit of salvolatile column.The proof-testing results indicated that the model possess fine predicting ability. |
| |
| Keywords: | salvolatile column carbonization degree ammonia concentration general regression neural net- works genetic algorithm |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
