| 基于神经网络的催化裂解产品产率模型 |
| |
| 引用本文： | 马法书,李真超,翁惠新.基于神经网络的催化裂解产品产率模型[J].炼油技术与工程,2001,31(2):35-38. |
| |
| 作者姓名： | 马法书 李真超 翁惠新 |
| |
| 作者单位： | 华东理工大学( |
| |
| 摘 要： | 采用神经网络方法 ,构造了一个催化裂解产品产率的BP神经网络 ,并利用Levenberg Marquardt算法来提高收敛速度及克服局部极值。模型预测结果液化石油气、汽油、柴油、丙烯和焦炭加损失产率的误差分别为 1.8% ,2 .4% ,5 .7% ,5 .8% ,6.3 % ,能够满足工业应用需求。
|
| 关 键 词： | 催化裂解 收率 神经网络 模拟仿真 |
| 修稿时间： | 2000年10月19 |
A MODEL OF DEEP CATALYTIC CRACKING PRODUCT YIELDS USING ARTIFICIAL NEURAL NETWORKS |
| |
| Ma Fashu,Li Zhenchao,Weng Huixin East China University of Science and Technology.A MODEL OF DEEP CATALYTIC CRACKING PRODUCT YIELDS USING ARTIFICIAL NEURAL NETWORKS[J].Petroleum Refinery Engineering,2001,31(2):35-38. |
| |
| Authors: | Ma Fashu Li Zhenchao Weng Huixin East China University of Science and Technology |
| |
| Affiliation: | Shanghai 200237 |
| |
| Abstract: | Based on artificial neural networks,a model predicting yields of DCC products was established.The model used Levenberg Marquardt algorithm to promote convergence and handle part extremism.The results showed that average errors of LPG,gasoline,diesel,propylene,coke plus loss are 1.8%,2.4%,5.7%,5.8%,6.3% respectively,meeting the commercial requirements. |
| |
| Keywords: | deep catalytic cracking yield neural networks analog simulation |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
