skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Weighted partial least squares method to improve calibration precision for spectroscopic noise-limited data

Abstract

Multivariate calibration methods have been applied extensively to the quantitative analysis of Fourier transform infrared (FT-IR) spectral data. Partial least squares (PLS) methods have become the most widely used multivariate method for quantitative spectroscopic analyses. Most often these methods are limited by model error or the accuracy or precision of the reference methods. However, in some cases, the precision of the quantitative analysis is limited by the noise in the spectroscopic signal. In these situations, the precision of the PLS calibrations and predictions can be improved by the incorporation of weighting in the PLS algorithm. If the spectral noise of the system is known (e.g., in the case of detector-noise-limited cases), then appropriate weighting can be incorporated into the multivariate spectral calibrations and predictions. A weighted PLS (WPLS) algorithm was developed to improve the precision of the analysis in the case of spectral-noise-limited data. This new PLS algorithm was then tested with real and simulated data, and the results compared with the unweighted PLS algorithm. Using near-infrared (NIR) calibration precision when the WPLS algorithm was applied. The best WPLS method improved prediction precision for the analysis of one of the minor components by a factor of nearly 9 relative tomore » the unweighted PLS algorithm.« less

Authors:
;
Publication Date:
Research Org.:
Sandia National Labs., Albuquerque, NM (United States)
Sponsoring Org.:
USDOE, Washington, DC (United States)
OSTI Identifier:
531089
Report Number(s):
SAND-97-2226C; CONF-970812-1
ON: DE97009353; TRN: 97:005281
DOE Contract Number:
AC04-94AL85000
Resource Type:
Conference
Resource Relation:
Conference: 11. international conference on Fourier transform spectroscopy, Athens, GA (United States), 10-15 Aug 1997; Other Information: PBD: 1997
Country of Publication:
United States
Language:
English
Subject:
44 INSTRUMENTATION, INCLUDING NUCLEAR AND PARTICLE DETECTORS; 66 PHYSICS; 99 MATHEMATICS, COMPUTERS, INFORMATION SCIENCE, MANAGEMENT, LAW, MISCELLANEOUS; INFRARED SPECTRA; MULTIVARIATE ANALYSIS; QUANTITATIVE CHEMICAL ANALYSIS; LEAST SQUARE FIT; FOURIER TRANSFORM SPECTROMETERS; INFRARED SPECTROMETERS; CALIBRATION

Citation Formats

Haaland, D.M., and Jones, H.D.T. Weighted partial least squares method to improve calibration precision for spectroscopic noise-limited data. United States: N. p., 1997. Web.
Haaland, D.M., & Jones, H.D.T. Weighted partial least squares method to improve calibration precision for spectroscopic noise-limited data. United States.
Haaland, D.M., and Jones, H.D.T. 1997. "Weighted partial least squares method to improve calibration precision for spectroscopic noise-limited data". United States. doi:. https://www.osti.gov/servlets/purl/531089.
@article{osti_531089,
title = {Weighted partial least squares method to improve calibration precision for spectroscopic noise-limited data},
author = {Haaland, D.M. and Jones, H.D.T.},
abstractNote = {Multivariate calibration methods have been applied extensively to the quantitative analysis of Fourier transform infrared (FT-IR) spectral data. Partial least squares (PLS) methods have become the most widely used multivariate method for quantitative spectroscopic analyses. Most often these methods are limited by model error or the accuracy or precision of the reference methods. However, in some cases, the precision of the quantitative analysis is limited by the noise in the spectroscopic signal. In these situations, the precision of the PLS calibrations and predictions can be improved by the incorporation of weighting in the PLS algorithm. If the spectral noise of the system is known (e.g., in the case of detector-noise-limited cases), then appropriate weighting can be incorporated into the multivariate spectral calibrations and predictions. A weighted PLS (WPLS) algorithm was developed to improve the precision of the analysis in the case of spectral-noise-limited data. This new PLS algorithm was then tested with real and simulated data, and the results compared with the unweighted PLS algorithm. Using near-infrared (NIR) calibration precision when the WPLS algorithm was applied. The best WPLS method improved prediction precision for the analysis of one of the minor components by a factor of nearly 9 relative to the unweighted PLS algorithm.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = 1997,
month = 9
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • Multivariate calibration methods have been applied extensively to the quantitative analysis of Fourier transform infrared (FT-IR) spectral data. Partial least squares (PLS) methods have become the most widely used multivariate method for quantitative spectroscopic analyses. Most often these methods are limited by model error or the accuracy or precision of the reference methods. However, in some cases, the precision of the quantitative analysis is limited by the noise in the spectroscopic signal. In these situations, the precision of the PLS calibrations and predictions can be improved by the incorporation of weighting in the PLS algorithm. If the spectral noise ofmore » the system is known (e.g., in the case of detector-noise-limited cases), then appropriate weighting can be incorporated into the multivariate spectral calibrations and predictions. A weighted PLS (WPLS) algorithm was developed to improve the precision of the analyses in the case of spectral-noise-limited data. This new PLS algorithm was then tested with real and simulated data, and the results compared with the unweighted PLS algorithm. Using near-infrared (NIR) spectra obtained from a series of dilute aqueous solutions, the simulated data produced calibrations that demonstrate improved calibration precision when the WPLS algorithm was applied. The best WPLS method improved prediction precision for the analysis of one of the minor components by a factor of nearly 9 relative to the unweighted PLS algorithm. {copyright} {ital 1998 American Institute of Physics.}« less
  • The concept of calibration of a thermal-mechanical fuel-pin performance code is examined for the purpose of developing a definition of code calibration based on well-established statistical measures of goodness of fit to experimental data which are free of the subjective judgements of the analyst doing the calibration. To achieve this purpose the least-squares/maximum-likelihood method is formulated for inferring the expected values and convariance matrix of any chosen set of fuel-pin materials correlation parameters from irradiation data.
  • The psuedo univariate limit of detection was calculated to compare to the multivariate interval. ompared with results from the psuedounivariate LOD, the multivariate LOD includes other factors (i.e. signal uncertainties) and the reveals the significance in creating models that not only use the analyte’s emission line but also its entire molecular spectra.
  • Full-spectrum multivariate calibration methods are capable of providing a multitude of diagnostic capabilities for evaluating the quality of the calibration, identifying problem calibration samples, and flagging unknown samples whose analysis by these methods might be unreliable. These diagnostics are demonstrated for the analysis of BPSG thin films on silicon using infrared spectroscopy and partial least-squares methods. 3 refs., 1 fig.