Implementation of a generalized least-squares method for determining calibration curves from data with general uncertainty structures

The determination of a best-fit calibration curve that describes the response of a measuring system to the value of a standard is one of the most widely used procedures in metrology. The mathematical basis for a generalized least-squares solution to this problem is reviewed. Examples of the application of a software implementation of the method are presented to illustrate the treatment of calibration problems with different uncertainty structures for the calibration data, including correlated data. The examples concern the calibration of analysers to measure the composition of natural gas and the calibration of a gas flow dilutor.

[1]  Bernard Butler,et al.  OPTIMISATION ALGORITHMS FOR GENERALISED DISTANCE REGRESSION IN METROLOGY , 2000 .

[2]  Gergely Vargha,et al.  Harmonisation of coupled calibration curves to reduce correlated effects in the analysis of natural gas by gas chromatography. , 2005, Journal of chromatography. A.

[3]  Francesca Pennecchi,et al.  On the best fit of a line to uncertain observation pairs , 2005 .

[4]  A. B. Forbes,et al.  Generalised regression problems in metrology , 1993, Numerical Algorithms.

[5]  Wolfram Bremser,et al.  Analysis of natural gas by gas chromatography reduction of correlated uncertainties by normalisation. , 2004, Journal of chromatography. A.

[6]  W. Press,et al.  Numerical Recipes in C++: The Art of Scientific Computing (2nd edn)1 Numerical Recipes Example Book (C++) (2nd edn)2 Numerical Recipes Multi-Language Code CD ROM with LINUX or UNIX Single-Screen License Revised Version3 , 2003 .

[7]  C. W. Clenshaw,et al.  Curve and Surface Fitting , 1965 .

[8]  D. Zwick,et al.  Applications of orthogonal distance regression in metrology , 1997 .

[9]  G. J. Babu,et al.  Linear regression in astronomy. II , 1990 .

[10]  M. Milton,et al.  Validation of the gravimetric values and uncertainties of independently prepared primary standard gas mixtures , 2006 .

[11]  C. Paige.  Fast Numerically Stable Computations for Generalized Linear Least Squares Problems , 1979 .

[12]  P. Boggs,et al.  A Stable and Efficient Algorithm for Nonlinear Orthogonal Distance Regression , 1987 .