1、Designation:D832122Standard Practice forDevelopment and Validation of Multivariate Analyses for Usein Predicting Properties of Petroleum Products,LiquidFuels,and Lubricants based on SpectroscopicMeasurements1This standard is issued under the fixed designation D8321;the number immediately following t
2、he designation indicates the year oforiginal 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.1.Scope*1.1 This practice covers a guide
3、for the multivariate cali-bration of infrared(IR)spectrophotometers and Raman spec-trometers used in determining the physical,chemical,andperformance properties of petroleum products,liquid fuelsincluding biofuels,and lubricants.This practice is applicable toanalyses conducted in the near infrared(N
4、IR)spectral region(roughly 780 nm to 2500 nm)through the mid infrared(MIR)spectral region(roughly 4000 cm-1to 40 cm-1).For Ramananalyses,this practice is generally applied to Stokes shiftedbands that occur roughly 400 cm-1to 4000 cm-1below thefrequency of the excitation.NOTE1While the practice descr
5、ibed herein deals specifically withmid-infrared,near-infrared,and Raman analysis,much of the mathemati-cal and procedural detail contained herein is also applicable for multivari-ate quantitative analysis done using other forms of spectroscopy.The useris cautioned that typical and best practices for
6、 multivariate quantitativeanalysis using other forms of spectroscopy may differ from the practicedescribed herein for mid-infrared,near-infrared,and Raman spectrosco-pies.1.2 Procedures for collecting and treating data for develop-ing IR and Raman calibrations are outlined.Definitions,terms,and cali
7、bration techniques are described.The calibration es-tablishes a multivariate correlation between the spectral fea-tures and the properties to be predicted.This correlation isherein referred to as the multivariate model.Criteria forvalidating the performance of the multivariate model aredescribed.The
8、 properties against which a multivariate model iscalibrated and validated are measured by Primary Test Meth-ods(PTMs)and the results of the PTM measurement are hereinreferred to as Primary Test Method Results(PTMR).Theanalysis of the spectra using the multivariate model produces aPredicted Primary T
9、est Method Result(PPTMR).1.3 The implementation of this practice requires that the IRspectrophotometer or Raman spectrometer has been installed incompliance with the manufacturers specifications.In addition,it assumes that,at the time of calibration,validation,andanalysis,the analyzer is operating a
10、t the conditions specified bythe manufacturer.The practice includes instrument perfor-mance tests which define the instrument performance at thetime of calibration,and which qualify the instrument bydemonstrating comparable performance during validation andanalysis.1.4 This practice covers technique
11、s that are routinely ap-plied for online,at-line,and laboratory quantitative analysis.The practice outlined covers the general cases for liquids andsolids that are single phase homogeneous samples whenpresented to the analyzers.Online application is limited bysample viscosity and the ability to intr
12、oduce sample to theanalyzer.All techniques covered require the use of a computerfor data collection and analysis.1.5 This practice is most typically applied when the spectraand the PTMR against which the analysis is calibrated aremeasured on the same sample.However,for some applications,spectra may
13、be measured on a basestock and the PTMR may bemeasured on the same basestock after constant level additiva-tion.1.5.1 Biofuel applications will typically fall into threecategories.1.5.1.1 The spectra and the PTM both measure the finishedbiofuel blend.1.5.1.2 The spectra are measured on a petroleum d
14、erivedblendstock,and the PTM measures the same blendstock after aconstant level additivation with the biocomponent.1.5.1.3 The spectra and PTM both measured the petroleumderived blendstock,and the PPTMRs from the multivariatemodel are used as inputs into a second model which predictsthe results obta
15、ined when the PTM is applied to the analysis ofthe finished blended product.The practice described hereinonly applies to the first of these two models.1This practice is under the jurisdiction of ASTM Committee D02 on PetroleumProducts,Liquid Fuels,and Lubricants and is the direct responsibility of S
16、ubcom-mittee D02.25 on Performance Assessment and Validation of Process StreamAnalyzer Systems.Current edition approved April 1,2022.Published June 2022.Originallyapproved in 2020.Last previous edition approved in 2021 as D8321 21.DOI:10.1520/D8321-22.*A Summary of Changes section appears at the end of this standardCopyright ASTM International,100 Barr Harbor Drive,PO Box C700,West Conshohocken,PA 19428-2959.United StatesThis international standard was developed in accordance with internationall