Improving the prediction capability of radial basis function networks

被引：9

作者：

Gurumoorthy, A ^{[1
]}

Kosanovich, KA ^{[1
]}

机构：

[1] Univ S Carolina, Dept Chem Engn, Columbia, SC 29208 USA

来源：

INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH | 1998年 / 37卷 / 10期

关键词：

D O I：

10.1021/ie980278o

中图分类号：

TQ [化学工业];

学科分类号：

0817 ;

摘要：

Radial basis function networks (RBFNs) have been well established as a class of supervised neural networks. Because they do not require a priori process information and they possess localized interpolation properties, they are attractive for the empirical modeling of complex nonlinear multivariable processes such as those associated with the chemical industry. In this paper, classical regularization theory is used to develop a technique for improving the prediction capabilities of RBFNs by incorporating process knowledge obtained from physicochemical relationships through modification of the objective function. An analysis of this new objective function is provided and compared to the conventional least-squares objective function. Several chemical process examples are provided to demonstrate the improved predictive capability of this modified RBFN.

引用

页码：3956 / 3970

页数：15

共 36 条

[21]

LEONARD JA, 1991, IEEE CONTROL SYS APR, P31

[22]

LEONARD JA, 1992, COMPUT CHEM ENG, P31

[23]

Lippmann R. P., 1988, Computer Architecture News, V16, P7, DOI [10.1109/MASSP.1987.1165576, 10.1145/44571.44572]

[24]

MCAVOY TJ, 1989, JT C NEUR NETW, P1227

[25] INTERPOLATION OF SCATTERED DATA - DISTANCE MATRICES AND CONDITIONALLY POSITIVE DEFINITE FUNCTIONS [J].

MICCHELLI, CA .

CONSTRUCTIVE APPROXIMATION, 1986, 2 (01) :11-22

[26] NONLINEAR INTERNAL MODEL CONTROL STRATEGY FOR NEURAL NETWORK MODELS [J].

NAHAS, EP ;

HENSON, MA ;

SEBORG, DE .

COMPUTERS & CHEMICAL ENGINEERING, 1992, 16 (12) :1039-1057

[27]

Narendra K. S., 1989, LEARNING AUTOMATA IN

[28] NETWORKS FOR APPROXIMATION AND LEARNING [J].

POGGIO, T ;

GIROSI, F .

PROCEEDINGS OF THE IEEE, 1990, 78 (09) :1481-1497

[29] REGULARIZATION ALGORITHMS FOR LEARNING THAT ARE EQUIVALENT TO MULTILAYER NETWORKS [J].

POGGIO, T ;

GIROSI, F .

SCIENCE, 1990, 247 (4945) :978-982

[30] A HYBRID NEURAL NETWORK-1ST PRINCIPLES APPROACH TO PROCESS MODELING [J].

PSICHOGIOS, DC ;

UNGAR, LH .

AICHE JOURNAL, 1992, 38 (10) :1499-1511

← 1 2 3 4 →