[1] Huang Kai, Luo Zhenghong, Chen Fengqiu, et al. Application of nonlinear partial least squarein catalyst modeling [J]. Journal of Southeast University (English Edition), 2004, 20 (1): 65-69. [doi:10.3969/j.issn.1003-7985.2004.01.014]
Copy
Application of nonlinear partial least squarein catalyst modeling(
)
非线性偏最小二乘法在催化剂建模中的应用
)Journal of Southeast University (English Edition)[ISSN:1003-7985/CN:32-1325/N]
- Volumn:
- 20
- Issue:
- 2004 1
- Page:
- 65-69
- Research Field:
- Chemistry and Chemical Engineering
- Publishing date:
- 2004-03-30
Info
- Title:
- Application of nonlinear partial least squarein catalyst modeling
- 非线性偏最小二乘法在催化剂建模中的应用
- Author(s):
- Huang Kai1; 3; Luo Zhenghong2; Chen Fengqiu3; Lü Dewei3
-
1Department of Chemistry and Chemical Engineering, Southeast University, Nanjing 210096, China
2Department of Chemical Engineering, Xiamen University, Xiamen 361005, China
3Department of Chemical Engineering, Zhejiang University, Hangzhou 310027, China
- PACS:
- TQ426.6
- DOI:
- 10.3969/j.issn.1003-7985.2004.01.014
- Abstract:
- In this paper neural network partial least square(NNPLS)was used to establish a robust reaction model for a multi-component catalyst of methane oxidative coupling. The details, including the learning algorithm, the number of hidden units of the inner network, activation function, initialization of the network weights and the principal components, are discussed. The results show that the structural organizations of inner neural network are 1-10-5-1, 1-8-4-1, 1-8-5-1, 1-7-4-1, 1-8-4-1, 1-8-6-1, respectively. The Levenberg-Marquardt method was used in the learning algorithm, and the central sigmoidal function is the activation function. Calculation results show that four principal components are convenient in the use of the multi-component catalyst modeling of methane oxidative coupling. Therefore a robust reaction model expressed by NNPLS succeeds in correlating the relations between elements in catalyst and catalytic reaction results. Compared with the direct network modeling, NNPLS model can be adjusted by experimental data conveniently and the calculation of the model is simpler and faster than that of the direct network model.
- 神经网络偏最小二乘法(NNPLS)被应用于一种甲烷氧化偶联多组分催化剂的鲁棒反应模型的建立.重点研究了内层神经网络学习算法、激活函数、网络结构(包括隐含节点数、隐含层)、网络权值初始化及主元的选取原则等.研究表明, 内层神经网络分别采用1-10-5-1, 1-8-4-1, 1-8-5-1, 1-7-4-1, 1-8-4-1, 1-8-6-1的拓扑结构是合适的; Levenberg-Marquardt方法被用于网络的学习算法可以加快学习速度; 同时采用了sigmoid函数为激活函数.计算结果显示, 四主元可以满足建模的需要.与单纯的神经网络催化剂模型相比, NNPLS方法压缩分解了变量, 减少了计算量, 同时使模型的推广能力得到提高, 有效地改善了直接神经网络建模过程中催化剂模型泛化能力较差的缺点.
References:
[1] Wold H. Nonlinear estimation by iterative least squares procedures [A]. In: David F, ed. Research Papers in Statistics [C]. New York: John Wiley & Sons, 1966.411-444.
[2] Werbos P J. Beyond regression: new tools for pre￣diction and analysis in the behavioral science [D]. Cambridge, MA: Harvard University, 1974.
[3] Rumelhart D, Hinton G, Williams R. Parallel dis￣tri￣buted processing [M]. Cambridge, MA: MIT Press, 1986.
[4] Hornik K M, Stinchcombe M, White H. Multilayer feed-forward neural networks are universal approximators [J]. Neural Networks, 1989, 2(2): 359-366.
[5] Qin S J, McAvoy T J. Nonlinear PLS modeling using neural networks [J]. Computers Chem Engng, 1992, 16(4): 379-391.
[6] Holcomb T R, Morari M. PLS/neural networks [J]. Computers Chem Engng, 1992, 16(4): 393-411.
[7] Huang K, Chen F Q, Lü D W. Artificial neural network-aided design of a multi-component catalyst for methane oxidative coupling [J]. Appl Catal A, 2001, 219: 61-68.
[8] Huang K. Research on computer-aided catalyst design and an application in multi-component catalyst design of methane oxidative coupling [D]. Hangzhou: Department of Chemical Engineering, Zhejiang University, 2001.
[9] Geladi P, Kowalski B R. An example of 2-block pre￣dic￣tive partial least-squares regression with simulated data [J]. Analyt Chim Acta, 1986, 185: 19-32.
Memo
- Memo:
- Biography: Huang Kai(1973—), male, doctor, lecturer, huangk@seu.edu.cn.