【国外标准】 Standard Practice for Performance-Based Qualification of Spectroscopic Analyzer Systems

5.1 This practice is intended for use by parties interested in releasing product by use of vibrational spectroscopic analyzer systems. It is expected to meet the industry need for a written practical reference describing a scientifically systematic approach to show the degree of confidence and degree of uncertainty in analyzer predicted values in relation to the PTM.5.2 This is a performance-based practice that relies on the demonstrated quality of the test result and on strict adherence to the referenced standards and the additional requirements in this practice.5.3 As part of demonstrating performance, this practice incorporates by reference other ASTM standardized practices as key steps in the process.5.4 There are prescriptive requirements included for this practice.5.4.1 The practice requires sample temperature to be carefully controlled in analyzer system hardware or that effects of temperature change be compensated in modeling or software.5.4.2 Outlier detection capability is required for demonstrating the multivariate calibration model is applicable for the analysis of the sample spectrum, that is, that the analysis interpolates the model, that the sample does not contain a statistically significant amount of unmodeled components above a certain limit based on spectral residual statistic and that the sample spectrum does not fall within gap in the multivariate calibration space.5.5 In order to follow this practice, all criteria must be met.5.5.1 The user shall investigate the cause of not meeting the practice requirements.5.5.2 For any nonconformities noticed, the user shall make corrections to the analyzer system or procedures to conform to the requirements of this practice.1.1 This practice covers requirements for establishing performance-based qualification of vibrational spectroscopic analyzer systems intended to be used to predict the test result of a material that would be produced by a Primary Test Method (PTM) if the same material is tested by the PTM.1.1.1 This practice provides methodology to establish the lower/upper prediction limits associated with the Predicted Primary Test Method Result (PPTMR) in 1.1 with a specified degree of confidence that would contain the PTM result (if tested by the PTM).1.1.2 The prediction limits in 1.1.1 can be used to estimate the confidence that product released using the analyzer system based on a PPTMR that meets PTM-based specification limits will meet PTM-based specification limits when tested by a PTM.1.2 The practice covers the qualification of on-line, at-line, or laboratory infrared or Raman analyzers used to predict physical, chemical, or performance properties of liquid petroleum products and fuels. Infrared analyzers can operate in the near-infrared (NIR) region, mid-infrared (MIR) region, or both.1.2.1 This practice applies to all analyzer systems that can meet the performance requirements defined within.1.2.2 This practice is not limited to analyzers designed by any specific instrument manufacturer.1.2.3 This practice allows for multiple calibration techniques to create a multivariate model which relates the spectra produced by the analyzer to the corresponding property determined by a PTM. Spectra can be used to predict multiple properties, but the analyzer system performance of each predicted property is qualified individually.1.3 The practice describes procedures for establishing performance requirements for analyzer system applications. The user of this practice must establish written protocols to confirm the procedures are being followed.1.4 This practice makes use of standard practices, guides, and methods already established in ASTM. Additional requirements are listed within this practice.1.5 Any multivariate model that meets performance requirements and detects when the spectrum of a sample is an outlier (analysis that represents an extrapolation of the model) or a nearest neighbor distance inlier (a spectrum residing in a gap in the multivariate space) can be used.1.6 This practice can be used with methods for determining properties of biofuel blends. Three alternative procedures can be used. In all three cases, the qualification of the predicted values for the blend are established and monitored as part of a continual program by application of Practice D6122 or by combined application of Practices D6122 and D3764 (see definition in section 3.1.18).1.6.1 If the analyzer is used to directly predict a property of the biofuel blend, and both the Primary Test Method Result (PTMR) and Predicted Primary Test Method Result (PPTMR) are measured on the same material, then the analyzer is validated using Practice D6122.1.6.2 If the analyzer is used to directly predict a property of a blend stock to which a fixed level of biofuel material is added prior to measurement by the PTM, and if the multivariate model correlates the spectrum of the blend stock to the PTMR for the fixed level blend, then the analyzer is validated using Practice D6122.1.6.3 If the analyzer directly predicts a property of a blend stock to which some amount of biofuel material is later added, then Practice D6122 is used to validate the analyzer performance. If the PPTMR produced by the analyzer is input into a second model to predict the property value for the final blend, based on the PPTMR for the blend stock and the blend level for the biofuel material, then the performance of this second model is validated using Practice D3764.1.7 Disclaimer of Liability as to Patented Inventions—Neither ASTM International nor an ASTM committee shall be responsible for identifying all patents under which a license is required in using this document. ASTM International takes no position respecting the validity of any patent rights asserted in connection with any item mentioned in this standard. Users of this standard are expressly advised that determination of the validity of any such patent rights, and the risk of infringement of such rights, are entirely their own responsibility.1.8 This standard does not purport to address all of the safety concerns, if any, associated with its use. It is the responsibility of the user of this standard to establish appropriate safety, health, and environmental practices and determine the applicability of regulatory limitations prior to use.1.9 This international standard was developed in accordance with internationally recognized principles on standardization established in the Decision on Principles for the Development of International Standards, Guides and Recommendations issued by the World Trade Organization Technical Barriers to Trade (TBT) Committee.