| 重整产品液收率软测量 |
| |
| 引用本文： | 孙自强,顾幸生,俞金寿.重整产品液收率软测量[J].华东理工大学学报(自然科学版),2002(Z1). |
| |
| 作者姓名： | 孙自强 顾幸生 俞金寿 |
| |
| 作者单位： | 华东理工大学自动化研究所 上海200237
(孙自强，顾幸生)，华东理工大学自动化研究所 上海200237(俞金寿) |
| |
| 基金项目： | 教育部高等学校骨干教师资助计划项目，上海市曙光计划资助项目 |
| |
| 摘 要： | 针对连续重整反应器 ,分别采用改进算法的 BP神经网络、PLS- BP神经网络和小波神经网络建立重整产品液收率软测量模型。经比较后发现小波神经网络所建模型拟合及预测效果最好 ,因此将其用于实际生产装置。使用结果表明该模型可以实时准确预测重整产品液收率。
|
| 关 键 词： | 模型 神经网络 BP PLS-BP 小波 |
Soft-sensing of Liquid Yield of Reforming Products |
| |
| SUN Zi qiang,GU Xing sheng ,YU Jin shou.Soft-sensing of Liquid Yield of Reforming Products[J].Journal of East China University of Science and Technology,2002(Z1). |
| |
| Authors: | SUN Zi qiang GU Xing sheng YU Jin shou |
| |
| Affiliation: | SUN Zi qiang,GU Xing sheng *,YU Jin shou |
| |
| Abstract: | Three soft sensing models based on improved BP artificial neural networks(ANN), PLS BP ANN and wavelet ANN respectively, have been established for liquid yield of reforming products. It can be seen after making a comparison among them that the model based on wavelet ANN offers the best performance. Therefore, it has been put into use for continuous reforming reactor and the testing results indicate that the liquid yield can be predicted real time and correctly. |
| |
| Keywords: | model neural network BP PLS BP wavelet |
| 本文献已被 CNKI 等数据库收录！ |
