POTENTIAL-ENERGY SURFACES FOR MACROMOLECULES - A NEURAL NETWORK TECHNIQUE

被引：82

作者：

SUMPTER, BG ^{[1
]}

NOID, DW ^{[1
]}

机构：

[1] UNIV TENNESSEE, DEPT CHEM, KNOXVILLE, TN 37996 USA

来源：

CHEMICAL PHYSICS LETTERS | 1992年 / 192卷 / 5-6期

基金：

美国国家科学基金会;

关键词：

D O I：

10.1016/0009-2614(92)85498-Y

中图分类号：

O64 [物理化学（理论化学）、化学物理学];

学科分类号：

070304 ; 081704 ;

摘要：

A method for obtaining potential energy surfaces for macromolecules is described. The basis of the method is the use of a neural network to learn the relationship between vibrational spectra and a multidimensional potential energy surface (PES). The results demonstrate that the neural network is capable of mapping the vibrational motion determined from spectra onto a fully coupled PES with relatively high levels of accuracy.

引用

页码：455 / 462

页数：8

共 33 条

[1] BAFFES PT, 1989, NNETS PROGRAM
[2] BARNES J, 1988, J PHYS REF DATA, V21, P194
[3] STUDY OF SKELETAL VIBRATIONS USED TO COMPUTE HEAT-CAPACITIES OF SOLID MACROMOLECULES
BU, HS
CHENG, SZD
WUNDERLICH, B
[J]. JOURNAL OF PHYSICAL CHEMISTRY, 1987, 91 (15) : 4179 - 4188
[4] CAUDILL M, 1989, AI EXPERT DEC, P46
[5] CAUDILL M, 1991, AI EXPERT JAN, P56
[6] CAUDILL M, 1988, AI EXPERT FEB, P55
[7] Cybenko G., 1989, Mathematics of Control, Signals, and Systems, V2, P303, DOI 10.1007/BF02551274
[8] Cybenko G., 1988, TECHNICAL REPORT
[9] HEHRE WJ, 1986, AB INITIO METHODS QU
[10] Hertz J., 1991, INTRO THEORY NEURAL

← 1 2 3 4 →