应用BP神经网络的二次反应清洁汽油辛烷值预测

周小伟 1; 袁俊 2; 杨伯伦 1

您当前的位置：

首页 >

文章列表页 >

应用BP神经网络的二次反应清洁汽油辛烷值预测

更新时间：2025-07-09

- 应用BP神经网络的二次反应清洁汽油辛烷值预测
- Prediction of Octane Number for Clean Gasoline Obtained from Secondary Reactions Based on Back-Propagation Neural Network
- 西安交通大学学报 2010年44卷第12期页码：82-86
- 作者机构：
  
  西安交通大学能源与动力工程学院,西安,710049
- 作者简介：
- 基金信息：
  
  : 2010-07-09. 作者简介: 周小伟(1984-),男,博士生;杨伯伦(联系人),男,教授,博士生导师. 基金项目: 国家“973计划”资助项目(2009CB219906);国家自然科学基金资助项目(20776117,20976144);高等学校博士学科点专项科研基金资助项目(20070698037)
- DOI：
  中图分类号： TQ622;TP301.6
- 网络首发：2010-12-10，
  
  纸质出版：2010
- 稿件说明：
移动端阅览
周小伟 1, 袁俊 2, 杨伯伦 1. 应用BP神经网络的二次反应清洁汽油辛烷值预测[J]. 西安交通大学学报, 2010,44(12):82-86.

Prediction of Octane Number for Clean Gasoline Obtained from Secondary Reactions Based on Back-Propagation Neural Network[J]. 2010, 44(12): 82-86.
周小伟 1, 袁俊 2, 杨伯伦 1. 应用BP神经网络的二次反应清洁汽油辛烷值预测[J]. 西安交通大学学报, 2010,44(12):82-86. DOI：

Prediction of Octane Number for Clean Gasoline Obtained from Secondary Reactions Based on Back-Propagation Neural Network[J]. 2010, 44(12): 82-86. DOI：

摘要

借鉴复杂反应动力学研究中的集总方法

将汽油辛烷值看成汽油链烷烃集总、环烷烃集总、芳烃集总、烯烃集总的函数.采用多元线性回归和BP神经网络算法

分别建立了二次反应清洁汽油的研究法辛烷值预测模型

并进行了实例计算验证和对比分析.结果表明

BP神经网络模型的整体性能优于多元线性回归模型

其强大的非线性映射能力能够更好地反映汽油研究法辛烷值与各集总组分之间的复杂关系

且具有更好的预测性能

模型预测值与实验测得的汽油辛烷值的平均相对误差为0.39%

与文献报道的汽油辛烷值的平均相对误差为0.92%.

Abstract

The octane number of gasoline was considered as a function of paraffin lump

naphthalene lump

aromatics lump and olefins lump based on lumping concept for complex reaction kinetics. Back-propagation(BP)neural network and multiple linear regression were adopted to establish prediction models for the research octane number(RON)of clean gasoline obtained from secondary reactions

respectively. The two models were compared and verified via several cases. The results show that BP neural network model exhibits better performance than multiple linear regression model for the higher prediction accuracy due to strong nonlinear mapping ability to reflex the complex relationship between RON and lump components. The mean absolute relative error between predicted gasoline RON and experimental data gets 0.39%

and 0.92% for the reported results.

关键词

Keywords

references

钱伯章, 朱建芳. 我国清洁汽油发展趋势及生产技术进展[J]. 天然气与石油, 2008, 26(6): 35-42.

QIAN Bozhang, ZHU Jianfang. Gasoline development trend and production technique progress in China[J]. Natural Gas and Oil, 2008, 26(6): 35-42.

YANG Bolun, ZHOU Xiaowei, CHEN Chun, et al. Molecule simulation for the secondary reactions of fluid catalytic cracking gasoline by the method of structure oriented lumping combined with Monte Carlo[J]. Industrial Engineering Chemistry Research, 2008, 47(14): 4648-4657.

YANG Bolun, ZHOU Xiaowei, YANG Xiaohui, et al. Multi-scale study on the secondary reactions of fluid catalytic cracking gasoline[J]. AIChE Journal, 2009, 55(8): 2138-2149.

李为民, 徐春明, 许志伟. 基于红外光谱-神经网络预测汽油辛烷值[J]. 炼油技术与工程, 2005, 35(12): 45-47.

LI Weimin, XU Chunming, XU Zhiwei. Predicting gasoline octane number with infrared spectrum-artificial neural network[J]. Petroleum Refinery Engineering, 2005, 35(12): 45-47.

Kelly J J, Barlow C H, Jinguji T M, et al. Prediction of gasoline octane numbers from near-infrared spectral features in the range 660～1215 nm[J]. Analytical Chemistry, 1989, 61(4): 313-320.

覃旭松, 戴连奎. 小波变换在Raman汽油辛烷值测定仪中的应用 [J].化工自动化及仪表, 2004, 31(5): 65-68.

QIN Xusong, DAI Liankui. Application of wavelet transform in gasoline octane number analyzer using Raman spectroscopy[J]. Control and Instruments in Chemical Industry, 2004, 31(5): 65-68.

李长秀, 杨海鹰, 王征. 一种新的汽油辛烷值的气相色谱测定方法[J]. 色谱, 2003, 21(1): 81-84.

LI Changxiu, YANG Haiying, WANG Zheng. A new method for determination of octane number of gasoline by chromatography[J]. Chinese Journal of Chromatography, 2003, 21(1): 81-84.

袁俊, 周小伟, 杨伯伦. 基于LM/SVM方法的二次反应清洁汽油辛烷值预测[J]. 高校化学工程学报, 2010, 24(2): 258-262.

YUAN Jun, ZHOU Xiaowei, YANG Bolun. Prediction of octane number for clean gasoline obtained from secondary reaction based on LM/SVM approach[J]. Journal of Chemical Engineering of Chinese Universities, 2010, 24(2):258-262.

WEEKMAN V W, NACE D M. Kinetic of catalytic cracking selectivity in fixed, moving and fluid bed reactors[J]. AIChE Journal, 1970, 16(3): 397-401.

朱晓明, 程建, 刘治国, 等. 基于反向传播算法神经网络的信用评分系统预测力研究[J]. 西安交通大学学报, 2006, 40(12): 1405-1409.

ZHU Xiaoming, CHENG Jian, LIU Zhiguo, et al. Study on prediction ability of credit scoring system based on neural network using back propagating algorithm[J]. Journal of Xi'an Jiaotong University, 2006, 40(12): 1405-1409.

蒋宗礼. 人工神经网络导论[M]. 北京:高等教育出版社, 2003.

ABILOV A, ZEYBEK Z. Use of neural network for modeling of non-linear process integration technology in chemical engineering[J]. Chemical Engineering and Processing, 2000, 39(5): 449-458.

高金森, 徐春明, 白跃华. 催化裂化汽油催化改质降烯烃反应规律的试验研究[J]. 炼油技术与工程, 2004, 34(5): 11-15.

GAO Jinsen, XU Chunming, BAI Yuehua. Test and study on the reaction rule of FCC naphtha olefin decrement upgrading[J]. Petroleum Refinery Engineering, 2004, 34(5): 11-15.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

一种基于神经网络的中制导改进算法

结合传感器阵列和BP神经网络的挥发性有机物定量检测系统

电子级超纯氮深冷空分流程节能优化研究

语义区分和子图整合驱动的知识图谱表示学习方法

超临界水异常区域热力学特性研究