核Hebbian算法在加氢脱芳烃过程中的建模应用

化工学报 ›› 2007, Vol. 58 ›› Issue (6): 1518-1522.

核Hebbian算法在加氢脱芳烃过程中的建模应用

王海清;宋执环;李平

浙江大学工业控制技术国家重点实验室

出版日期:2007-06-05 发布日期:2007-06-05

Modified kernel Hebbian algorithm with application to modeling of hydro-dearomatization process

WANG Haiqing,SONG Zhihuan,LI Ping

Online:2007-06-05 Published:2007-06-05

摘要/Abstract

摘要：

提出一种采用改进核Hebbian算法的加氢脱芳烃过程的递推产品质量建模方法;用于实时估计终端分馏产品的质量指标。通过利用核Hebbian算法的中间结果;计算中心化的核矩阵特征值;进而由核主元回归方法得到非线性动态质量模型。该递推滑动窗建模方法无需计算和保存整个核矩阵;并验证了所得到的闪点模型在正常和故障工况下均具有足够的精度。

关键词:

加氢脱芳烃, 产品质量建模, 统计学习理论, Hebbian算法

Abstract:

A modified kernel Hebbian algorithm (MKHA) was proposed to integrate with the kernel principal component regression (KPCR) method for recursive product quality modeling of a two-stage hydro-dearomatization process.The approach to calculating the eigenvalues of centering kernel matrix was derived and the whole algorithm is formulated in a recursive mode.The proposed modeling strategy has an advantage of no need to calculate and store the kernel matrix.The obtained recursive nonlinear dynamic flash point model showed satisfying precision under both normal and faulty operations, and comparison studies with traditional offline KPCR modeling were presented.

Key words:

加氢脱芳烃, 产品质量建模, 统计学习理论, Hebbian算法

王海清, 宋执环, 李平. 核Hebbian算法在加氢脱芳烃过程中的建模应用 [J]. 化工学报, 2007, 58(6): 1518-1522.

WANG Haiqing, SONG Zhihuan, LI Ping. Modified kernel Hebbian algorithm with application to modeling of hydro-dearomatization process[J]. CIESC Journal, 2007, 58(6): 1518-1522.

[1]	王海清, 蒋宁. 基于Kernel特征空间分解的组分仪递推模型 [J]. 化工学报, 2008, 59(1): 142-147.
[2]	王海清, 蒋宁. 用于过程故障诊断的自适应kernel学习网络分类器 [J]. 化工学报, 2007, 58(9): 2276-2280.
[3]	王海清蒋宁. 自适应Kernel学习网络在TE过程组分仪建模中的应用 [J]. CIESC Journal, 2007, 58(2): 425-430.