CONSTRAINED BACKGROUND BILINEARIZATION

被引：18

作者：

LIANG, YZ ^{[1
]}

MANNE, R ^{[1
]}

KVALHEIM, OM ^{[1
]}

机构：

[1] UNIV BERGEN,DEPT CHEM,N-5007 BERGEN,NORWAY

来源：

CHEMOMETRICS AND INTELLIGENT LABORATORY SYSTEMS | 1992年 / 14卷 / 1-3期

关键词：

D O I：

10.1016/0169-7439(92)80102-A

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

A new constrained optimization method is presented for background bilinearization of two-way multicomponent data. Constraints come from the positivity of spectral intensities and concentration. The convergence properties of the proposed method, called constrained background bilinearization, is tested by a series of simulations. The results compare favourably with a modified version of residual bilinearization, mainly because the constrained method is superior in locating the global optimum in the presence of local optima.

引用

页码：175 / 184

页数：10

共 19 条

[1]

AUNE B, 1991, 2ND SCAN S CHEM BERG

[2]

Avriel M, 2003, NONLINEAR PROGRAMMIN

[3] AN EXTENSION OF THE MULTIVARIATE COMPONENT-RESOLUTION METHOD TO 3 COMPONENTS [J].