Method and arrangement for automatic calibrating of spectrometers

• US 8,543,348 B2
• Filed: 06/03/2009
• Issued: 09/24/2013
• Est. Priority Date: 06/10/2008
• Status: Active Grant

• Current Assignee: Endress + Hauser Conducta Gesellschaft Fur Mess- Und Regeltechnik Mbh + Co. Kg
• Original Assignee: Endress + Hauser Conducta Gesellschaft Fur Mess- Und Regeltechnik Mbh + Co. Kg
• Inventors: Alber, Thomas, Andelic, Edin
• Primary Examiner(s): Lau, Tung S
• Application Number:
  US12/996,668
• Publication Number:
  US 20110093232A1
• Time in Patent Office: 1,574 Days
• Field of Search: 702/104
• US Class Current: 702/104
• CPC Class Codes: G01J 3/28 Investigating the spectrum using colour filters G01J3/51

Abstract

An arrangement and method for calibrating a spectrometer with a spectrometric apparatus for measuring spectral fractions of the substance to be examined as well as with a calibration unit. The calibrating of the spectrometric apparatus and/or measurement data based on measured spectrometric data is performed, by the following steps: a) measuring a totality of spectral fractions of the substance to be examined; b) storing the totality of the measured spectral fractions in a memory module in the form of a multidimensional coefficient vector; c) automatic extracting, from the totality of the measured spectral fractions physically relevant for the particular measuring procedure via an automatically running, iterative, estimation method furnished in the calibration unit; and d) calibrating the measured spectrometric measurement data for the investigated substance based on the step c) extracted, spectral fractions of the spectrometric data physically relevant for the substance to be examined.

Claims

1. The arrangement for calibrating a spectrometer for application in chemometrics or chemometric processes in the context of measuring spectral fractions of a substance to be examined, comprising:

a spectrometric apparatus for measuring spectral fractions X of the substance to be examined;

a calibration unit for calibrating the spectrometric apparatus;

a memory module for storing measured and evaluated, spectrometric data; and

a selection module, via which, based on a mathematical iteration method, physically relevant spectral fractions Xref of a measurement are automatically extractable, and said selection module is connected with said memory module and said spectrometric apparatus;

said arrangement is adapted to store the totality of the measured spectral fractions X in said memory module as spectrometric measurement data in the form of a multidimensional coefficient vector; and

said mathematical iterative estimation method comprises an iterative estimation method according to Bayes, via which coefficient values of the multidimensional coefficient vector tending toward zero during the iteration are removed before a regression analysis.

2. The arrangement as claimed in claim 1, wherein:

said selection module is connected via bidirectional data lines with said memory module and/or said spectrometric apparatus.

3. The arrangement as claimed in claim 1, wherein:

the mathematical iteration method in said selection module is furnished as a program, which is automatically executed in said calibration unit after each measurement.

4. The arrangement as claimed in claim 1, wherein:

the mathematical iteration algorithm in said selection module is based on a method applying Bayesian inference.

5. The arrangement as claimed in claim 1, further comprising:

a memory module, in which spectrometric reference data for measured substances as well as, in each case, relevant extracted spectral fractions X_rel of current measurements are stored and storable.

6. The arrangement as claimed in claim 1, further comprising:

a computing unit, by means of which, via comparison between stored spectrometric reference data and current measurement data calibration signals are producible, which can be input into said calibration unit of said spectrometric apparatus automatically for the calibrating.

7. The arrangement as claimed in claim 1, wherein:

said selection module is provided in said calibration unit;

said selection module is embodied to automatically extract said physically relevant spectral fractions X_rel from a totality of said spectral fractions X measured by said spectrometric apparatus based on said mathematical iteration method.

8. The arrangement as claimed in claim 1, wherein:

said physically relevant spectral fractions X_rel are those spectra or wavelengths in which said substance to be examined absorbs light.

9. The arrangement as claimed in claim 7, wherein:

said selection module is provided in said calibration unit;

said selection module is embodied to automatically extract said physically relevant spectral fractions X_rel from a totality of said spectral fractions X measured by said spectrometric apparatus based on said mathematical iteration method.

10. An arrangement for calibrating a spectrometer comprising:

a spectrometric apparatus for measuring spectral fractions X of a substance to be examined, said spectrometric apparatus being adapted to measure a totality of spectral fractions X of said substance;

a calibration unit for calibrating the spectrometric apparatus;

a memory module for storing measured and evaluated spectrometric data; and

a selection module provided in said calibration unit, said selection module being embodied to automatically extract physically relevant spectral fractions Xref from said totality of spectral fractions X measured by said spectrometric apparatus using a mathematical iterative estimation method and to store said physically relevant spectral fractions Xref in said memory module for calibrating the spectrometer arrangement;

said arrangement is adapted to store the totality of the measured spectral fractions X in said memory module as spectrometric measurement data in the form of a multidimensional coefficient vector; and

said mathematical iterative estimation method comprises an iterative estimation method according to Bayes, via which coefficient values of the multidimensional coefficient vector tending toward zero during the iteration are removed before a regression analysis.

11. The arrangement as claimed in claim 10, wherein:

said arrangement is further adapted to feed said physically relevant spectral fractions X_rel to the calibration unit for calibrating the spectrometric apparatus.

12. The arrangement as claimed in claim 10, wherein:

said mathematical iterative estimation method is based on a method applying Bayesian inference.

13. The arrangement as claimed in claim 10, wherein:

said physically relevant spectral fractions X_rel are those spectra or wavelengths in which said substance to be examined absorbs light.