1、Designation: D6122 13Standard Practice forValidation of the Performance of Multivariate Online, At-Line, and Laboratory Infrared Spectrophotometer BasedAnalyzer Systems1This standard is issued under the fixed designation D6122; the number immediately following the designation indicates the year ofor
2、iginal adoption or, in the case of revision, the year of last revision. A number in parentheses indicates the year of last reapproval. Asuperscript epsilon () indicates an editorial change since the last revision or reapproval.INTRODUCTIONOperation of a laboratory or process stream analyzer system t
3、ypically involves four sequentialactivities. (1) Analyzer CalibrationWhen an analyzer is initially installed, or after majormaintenance has been performed, diagnostic testing is performed to demonstrate that the analyzermeets the manufacturers specifications and historical performance standards. The
4、se diagnostic testsmay require that the analyzer be adjusted so as to provide predetermined output levels for certainreference materials. (2) CorrelationOnce the diagnostic testing is completed, process streamsamples are analyzed using both the analyzer system and the corresponding primary test meth
5、od(PTM). A mathematical function is derived that relates the analyzer output to the primary test method(PTM). The application of this mathematical function to an analyzer output produces a predictedprimary test method result (PPTMR). (3) Probationary ValidationOnce the relationship betweenthe analyz
6、er output and PTMRs has been established, a probationary validation is performed using anindependent but limited set of materials that were not part of the correlation activity. This probationaryvalidation is intended to demonstrate that the PPTMRs agree with the PTMRs to within user-specifiedrequir
7、ements for the analyzer system application. (4) General and Continual ValidationAfter anadequate number of PPTMRs and PTMRs have been accrued on materials that were not part of thecorrelation activity, a comprehensive statistical assessment is performed to demonstrate that thePPTMRs agree with the P
8、TMRs to within user-specified requirements. Subsequent to a successfulgeneral validation, quality assurance control chart monitoring of the differences between PPTMR andPTMR is conducted during normal operation of the process analyzer system to demonstrate that theagreement between the PPTMRs and th
9、e PTMRs established during the General Validation ismaintained. This practice deals with the third and fourth of these activities.1. Scope*1.1 This practice covers requirements for the validation ofmeasurements made by laboratory or process (online or at-line)near- or mid-infrared analyzers, or both
10、, used in the calculationof physical, chemical, or quality parameters (that is, properties)of liquid petroleum products. The properties are calculatedfrom spectroscopic data using multivariate modeling methods.The requirements include verification of adequate instrumentperformance, verification of t
11、he applicability of the calibrationmodel to the spectrum of the sample under test, and verificationof equivalence between the result calculated from the infraredmeasurements and the result produced by the primary testmethod used for the development of the calibration model.When there is adequate var
12、iation in property level, the statis-tical methodology of Practice D6708 is used to provide generalvalidation of this equivalence over the complete operatingrange of the analyzer. For cases where there is inadequateproperty variation, methodology for level specific validation isused.1.2 Performance
13、Validation is conducted by calculating theprecision and bias of the differences between results from theanalyzer system (or subsystem) produced by application of themultivariate model, (such results are herein referred to asPredicted Primary Test Method Results (PPTMRs), versus thePrimary Test Metho
14、d Results (PTMRs) for the same sampleset. Results used in the calculation are for samples that are not1This practice is under the jurisdiction of ASTM Committee D02 on PetroleumProducts and Lubricants and is the direct responsibility of Subcommittee D02.25 onPerformance Assessment and Validation of
15、Process Stream Analyzer Systems.Current edition approved May 1, 2013. Published June 2013. Originallyapproved in 1997. Last previous edition approved in 2010 as D6122 10. DOI:10.1520/D6122-13.*A Summary of Changes section appears at the end of this standardCopyright ASTM International, 100 Barr Harb
16、or Drive, PO Box C700, West Conshohocken, PA 19428-2959. United States1used in the development of the multivariate model. Thecalculated precision and bias are statistically compared touser-specified requirements for the analyzer system applica-tion.1.2.1 For analyzers used in product release or prod
17、uctquality certification applications, the precision and bias re-quirement for the degree of agreement are typically based onthe site or published precision of the Primary Test Method.NOTE 1In most applications of this type, the PTM is the specification-cited test method.1.2.2 This practice does not
18、 describe procedures for estab-lishing precision and bias requirements for analyzer systemapplications. Such requirements must be based on the critical-ity of the results to the intended business application and oncontractual and regulatory requirements. The user must estab-lish precision and bias r
19、equirements prior to initiating thevalidation procedures described herein.1.3 This practice does not cover procedures for establishingthe calibration model (correlation) used by the analyzer.Calibration procedures are covered in Practices E1655 andreferences therein.1.4 This practice is intended as
20、a review for experiencedpersons. For novices, this practice will serve as an overview oftechniques used to verify instrument performance, to verifymodel applicability to the spectrum of the sample under test,and to verify equivalence between the parameters calculatedfrom the infrared measurement and
21、 the results of the primarytest method measurement.1.5 This practice teaches and recommends appropriate sta-tistical tools, outlier detection methods, for determiningwhether the spectrum of the sample under test is a member ofthe population of spectra used for the analyzer calibration. Thestatistica
22、l tools are used to determine if the infrared measure-ment results in a valid property or parameter estimate.1.6 The outlier detection methods do not define criteria todetermine whether the sample or the instrument is the cause ofan outlier measurement. Thus, the operator who is measuringsamples on
23、a routine basis will find criteria to determine that aspectral measurement lies outside the calibration, but will nothave specific information on the cause of the outlier. Thispractice does suggest methods by which instrument perfor-mance tests can be used to indicate if the outlier methods arerespo
24、nding to changes in the instrument response.1.7 This practice is not intended as a quantitative perfor-mance standard for the comparison of analyzers of differentdesign.1.8 Although this practice deals primarily with validation ofinfrared analyzers, the procedures and statistical tests describedhere
25、in are also applicable to other types of analyzers whichemploy multivariate models.1.9 This standard does not purport to address all of thesafety concerns, if any, associated with its use. It is theresponsibility of the user of this standard to establish appro-priate safety and health practices and
26、determine the applica-bility of regulatory limitations prior to use.2. Referenced Documents2.1 ASTM Standards:2D1265 Practice for Sampling Liquefied Petroleum (LP)Gases, Manual MethodD3764 Practice for Validation of the Performance of ProcessStream Analyzer SystemsD4057 Practice for Manual Sampling
27、of Petroleum andPetroleum ProductsD4177 Practice for Automatic Sampling of Petroleum andPetroleum ProductsD6299 Practice for Applying Statistical Quality Assuranceand Control Charting Techniques to Evaluate AnalyticalMeasurement System PerformanceD6708 Practice for Statistical Assessment and Improve
28、mentof Expected Agreement Between Two Test Methods thatPurport to Measure the Same Property of a MaterialD7278 Guide for Prediction ofAnalyzer Sample System LagTimesD7453 Practice for Sampling of Petroleum Products forAnalysis by Process Stream Analyzers and for ProcessStream Analyzer System Validat
29、ionD7808 Practice for Determining the Site Precision of aProcess Stream Analyzer on Process Stream MaterialE131 Terminology Relating to Molecular SpectroscopyE275 Practice for Describing and Measuring Performance ofUltraviolet and Visible SpectrophotometersE456 Terminology Relating to Quality and St
30、atisticsE932 Practice for Describing and Measuring Performance ofDispersive Infrared SpectrometersE1421 Practice for Describing and Measuring Performanceof Fourier Transform Mid-Infrared (FT-MIR) Spectrom-eters: Level Zero and Level One TestsE1655 Practices for Infrared Multivariate QuantitativeAnal
31、ysisE1866 Guide for Establishing Spectrophotometer Perfor-mance TestsE1944 Practice for Describing and Measuring Performanceof Laboratory Fourier Transform Near-Infrared (FT-NIR)Spectrometers: Level Zero and Level One Tests3. Terminology3.1 Definitions:3.1.1 For definitions of terms and symbols rela
32、ting to IRspectroscopy, refer to Terminology E131.3.1.2 For definitions of terms and symbols relating tomultivariate calibration, refer to Practices E1655.3.1.3 For definitions of terms relating to statistical qualitycontrol, refer to Practice D6299 and Terminology E456.3.1.4 between-method reproduc
33、ibility (RXY), na quantita-tive expression of the random error associated with thedifference between two results obtained by different operatorsusing different apparatus and applying the two methods X andY, respectively, each obtaining a single result on an identical2For referenced ASTM standards, v
34、isit the ASTM website, www.astm.org, orcontact ASTM Customer Service at serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.D6122 132test sample, when the methods have been assessed and anappropriate bias-correction ha
35、s been applied in accordancewith this practice; it is defined as the 95 % confidence limit forthe difference between two such single and independentresults. D67083.1.5 control limits, nlimits on a control chart which areused as criteria for signaling the need for action, or for judgingwhether a set
36、of data does or does not indicate a state ofstatistical control. E4563.2 Definitions of Terms Specific to This Standard:3.2.1 action limit, nthe limiting value from an instrumentperformance test, beyond which the analyzer is expected toproduce potentially invalid results.3.2.2 analyzer, nall piping,
37、 hardware, computer, software,instrumentation and calibration model required to automati-cally perform analysis of a process or product stream.3.2.3 analyzer calibration, nsee multivariate calibration.3.2.4 analyzer site precision, na statistical measure of theexpected long-term variability of analy
38、zer results for sampleswhose spectra are neither outliers, nor nearest neighbor inliers.3.2.5 analyzer model, nsee multivariate model.3.2.6 analyzer repeatability, na statistical measure of theexpected short-term variability of results produced by theanalyzer for samples whose spectra are neither ou
39、tliers nornearest neighbor inliers.3.2.7 analyzer result, nthe numerical estimate of aphysical, chemical, or quality parameter produced by applyingthe calibration model to the spectral data collected by theanalyzer.3.2.8 analyzer validation test, nsee validation test.3.2.9 calibration transfer, na m
40、ethod of applying a mul-tivariate calibration developed on one analyzer to a differentanalyzer by mathematically modifying the calibration model orby instrument standardization.3.2.10 check sample, na single, pure liquid hydrocarboncompound or a known, reproducible mixture of liquid hydro-carbon com
41、pounds whose spectrum is constant over time suchthat it can be used in a performance test.3.2.11 exponentially weighted moving average controlchart, na control chart based on the exponentially weightedaverage of individual observations from a system; the obser-vations may be the differences between
42、the analyzer result, andthe result from the primary test method.3.2.12 individual observation control chart, na controlchart of individual observations from a system; the observa-tions may be the differences between the analyzer result andthe result from the primary test method.3.2.13 inlier, nsee n
43、earest neighbor distance inlier.3.2.14 inlier detection methods, nstatistical tests whichare conducted to determine if a spectrum resides within aregion of the multivariate calibration space, which is sparselypopulated.3.2.15 in-line probe, na spectrophotometer cell installedin a process pipe or sli
44、p stream loop and connected to theanalyzer by optical fibers.3.2.16 instrument, nspectrophotometer, associated elec-tronics and computer, spectrophotometer cell and, if utilized,transfer optics.3.2.17 instrument standardization, na procedure for stan-dardizing the response of multiple instruments su
45、ch that acommon multivariate model is applicable for measurementsconducted by these instruments, the standardization beingaccomplished by way of adjustment of the spectrophotometerhardware or by way of mathematical treatment of the collectedspectra.3.2.18 line sample, na process or product sample wh
46、ich iswithdrawn from a sample port in accordance with PracticesD1265, D4057, D4177,orD7453, whichever is applicable,during a period when the material flowing through the analyzeris of uniform quality and the analyzer result is essentiallyconstant.3.2.19 moving range of two control chart, na control
47、chartthat monitors the change in the absolute value of the differencebetween two successive differences of the analyzer resultminus the result from the primary test method.3.2.20 multivariate calibration, nan analyzer calibrationthat relates the spectrum at multiple wavelengths or frequen-cies to th
48、e physical, chemical, or quality parameters.3.2.21 multivariate model, na multivariate, mathematicalrule or formula used to calculate physical, chemical, or qualityparameters from the measured infrared spectrum.3.2.22 nearest neighbor distance inlier, na spectrum re-siding within a gap in the multiv
49、ariate calibration space, theresult for which is subject to possible interpolation error.3.2.23 optical background, nthe spectrum of radiationincident on a sample under test, typically obtained by measur-ing the radiation transmitted through the spectrophotometercell when no sample is present, or when an optically thin ornonabsorbing liquid is present.3.2.24 optical reference filter, nan optical filter or otherdevice which can be inserted into the optical path in thespectrophotometer or probe producing an absorption spectrumwhich is known to be const
