1、 EEMUA Code of Practice Calibration and validation of process analysers PUBLICATION 175 Edition 2 Released by IHS. NOT FOR RESALE THE ENGINEERING EQUIPMENT AND MATERIALS USERS ASSOCIATION EEMUA Code of Practice Calibration and validation of process analysers PUBLICATION 175 Edition 2 Copyright 2013
2、The Engineering Equipment and Materials Users Association. A company limited by guarantee. Registered in England. Company number 477838. ISBN 978 0 85931 197 7 Imprint reference 09-2013 Registered and Trading Address 63 Mark Lane London EC3R 7NQ Telephone: +44 (0)20 7488 0801 Fax: +44 (0)20 7488 349
3、9 E-mail: saleseemua.org Website: www.eemua.org EEMUA Publication 175 Calibration and validation of process analysers EEMUA ii ENGINEERING EQUIPMENT AND MATERIALS USERS ASSOCIATION About EEMUA The Engineering Equipment and Materials Users Association is usually known by its initials, EEMUA. EEMUA is
4、 governed by a Council of members. This Council is responsible for the policies, financing, membership, and strategy of the Association. Any organisation that is a substantial user or purchaser of engineering products is eligible to apply to become a member of EEMUA. Membership is subject to the app
5、roval of Council. EEMUA Aims EEMUA aims to improve the safety, environmental and operating performance of industrial facilities in the most cost-effective way, pursuing and promoting leadership in industrial asset management by sharing engineering experiences and expertise, and by the active, enligh
6、tened promotion of the distinct interests of users of engineering products. These aims are achieved by various means, including: Providing the organisation within which networking, information sharing and collaboration among users on non-competitive technical matters can take place. Influencing the
7、way written regulations are interpreted and applied in practice. Promoting and presenting members views, and encouraging the application of good sound engineering practices. Developing user guides, specifications, training and competency schemes. Facilitating members participation in national and in
8、ternational standards making. Influencing relevant national and European legislation and regulation. The income and resources of EEMUA are applied solely towards the promotion of the aims of the Association. No profit is transferred by way of dividend, bonus or otherwise to members. The work of the
9、Association is mainly carried out by members representatives on EEMUA Technical Committees, Working Groups, and Forums, supported by EEMUA staff. Incorporation, Membership and Other Information For the purposes of conducting its business, EEMUA is incorporated in England and Wales as a company, limi
10、ted by guarantee, not having share capital. All usual aspects of company management and business administration are dealt with by the EEMUA executive directors, including policy implementation and leadership of the executive team (the staff). A list of EEMUA publications for sale is given at the end
11、 of this Publication. The full list is also on the Associations website, including details of on-line shopping facilities. To enquire about corporate Membership, write to enquirieseemua.org or call +44 (0)20 7488 0801. EEMUA Publication 175 Calibration and validation of process analysers EEMUA iii A
12、BOUT THIS PUBLICATION Legal Aspects All rights, title and interest in this Publication belong to EEMUA. All rights are reserved. No part of this Publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means: electronic, mechanical, photocopying, recording or
13、 other, without the express prior written agreement of the EEMUA Chief Executive. Infringement of copyright is illegal and also reduces the Associations income thereby jeopardising its ability to fund the production of future publications. It has been assumed in the preparation of this Publication t
14、hat: the user ensures selection of those parts of its contents appropriate to the intended application; and that such selection and application is correctly carried out by appropriately qualified and competent persons for whose guidance this Publication has been prepared. EEMUA and the individual me
15、mbers of the Work Group that prepared this EEMUA Publication do not, and indeed cannot, make any representation or give any warranty or guarantee in connection with material contained herein. EEMUA and the individual members of the Work Group that prepared this EEMUA Publication expressly disclaim a
16、ny liability or responsibility for damage or loss resulting from the use or reliance on this Publication. Any recommendations contained herein are based on the most authoritative information available at the time of writing and on current good engineering practice, but it is essential for the user t
17、o take account of pertinent subsequent developments and/or legislation. Any person who encounters an inaccuracy or ambiguity when making use of this Publication is asked to notify EEMUA without delay so that the matter may be investigated and appropriate action taken. Consultation and Feedback EEMUA
18、 encourages constructive comments on this Publication from both Members of the Association and others. Comments should be sent on the standard feedback form, a copy of which is provided towards the end of this Publication. An electronic version of the feedback form is available from EEMUA - e-mail y
19、our request for a copy to enquirieseemua.org, telephone your request to +44 (0)20 7488 0801, or download the form from the EEMUA website at www.eemua.org. Submit comments by e-mail (preferred) or post. Comments are considered by the relevant EEMUA Technical Committees and may be incorporated in futu
20、re editions of this Publication. New editions are publicised on the EEMUA website EEMUA Publication 175 Calibration and validation of process analysers EEMUA iv EEMUA Publication 175 Calibration and validation of process analysers EEMUA v Contents Preface ix 1. Scope: calibration and validation 1 2.
21、 Definitions 3 2.1 Equipment . 3 2.2 Measurement . 3 2.3 Calibration and validation 4 2.4 Statistical terms . 5 3. General considerations: calibration and validation 9 4. Preconditions for analyser calibration and validation . 11 4.1 General . 11 4.2 Factory Acceptance Testing (FAT) . 11 4.3 Site in
22、stallation 11 4.4 Site Acceptance Testing (SAT) 12 4.5 Site precision . 12 4.6 Read-out instrumentation 12 5. Principles of initial calibration, correlation and initial validation . 13 5.1 Initial calibration and correlation . 13 5.2 Probationary/initial validation . 14 5.3 Analyser result and accep
23、ted reference value comparison methods. . 14 6. Principles of routine validation and calibration adjustment 15 6.1 General . 15 6.2 Analyser result and reference test/primary test method comparison techniques 16 6.2.1 Reference sample method 16 6.2.2 Paired sample method . 16 6.3 Routine validation
24、(scheduled check of continuing performance) 17 6.4 Calibration adjustment (adjustment, if required, to re-establish analyser results within performance limits) . 17 6.5 Routine validation and calibration adjustment intervals 18 6.6 Number of result values required for routine validation and calibrat
25、ion adjustment . 19 6.7 Routine validation and calibration adjustment of model based/chemometric process analysers . 20 7. Comparison methods: reference sample method and paired sample method . 21 7.1 Comparing analyser and reference method results. 21 7.2 Routine validation and calibration adjustme
26、nt using the reference sample method 21 7.2.1 Advantages and limitations 21 7.2.2 Choice of reference samples . 22 7.2.3 Testing of reference samples 23 7.3 Routine validation and calibration adjustment using the paired sample method 23 7.3.1 Advantages and limitations 23 7.3.2 Collecting paired sam
27、ple results 25 8. Evaluation of routine validation and calibration adjustment results: use of statistical tools and control charts 27 8.1 General . 27 8.2 Use of statistical tools . 27 8.3 Routine validation and calibration adjustment: use of control charts. 27 8.3.1 CUSUM approach 29 8.4 Control ch
28、art limits . 30 EEMUA Publication 175 Calibration and validation of process analysers EEMUA vi 8.5 Control chart limits derived from analyser reference sample or paired sample laboratory data 31 Appendix 1: Example interpretation of control chart readings 33 Appendix 2: Determination of control char
29、t limits by measuring standard deviations of differences . 35 References and Bibliography 37 EEMUA Publication 175 Calibration and validation of process analysers EEMUA vii Figures Figure 1 Example of a control charting diagram28 Figure 2 Examples of analyser results in a CUSUM chart29 Figure 3 Exam
30、ple of accurately distributed control chart reading.33 Figure 4 Example of biased control chart reading.33 Figure 5 Example of drifting control chart reading33 Figure 6 Example control chart reading; value outside warning limit.34 Figure 7 Example determination of control chart limits by measuring s
31、tandard deviations36 Tables Table 1 Sample validation uncertainty results.19 Table 2 Example of the control chart limits for a distillation analyser.35 EEMUA Publication 175 Calibration and validation of process analysers EEMUA viii EEMUA Publication 175 Calibration and validation of process analyse
32、rs EEMUA ix Preface This Code of Practice has been prepared by the EEMUA On-Line Analysers Committee. It replaces the First Edition of this Publication, previously named Calibration and checking of process analysers, which in turn replaced the following historical IP Codes of Practice relating to an
33、alysers: IP 340/92, 341/82, 347/82, 348/82, 349/82 and 353/82. Other associated EEMUA publications are: EEMUA 138 - Design and installation of on-line analyser systems. EEMUA 226 - Design and installation of on-line analyser systems - a guide to technical enquiry and bid evaluation. EEMUA 187 - Anal
34、yser systems - a guide to maintenance management. The word analyser has been used throughout this Publication to refer to instruments, variously known as on-line analysers, process stream analysers, process analysers, quality analysers, quality measuring instruments and process quality monitors. EEM
35、UA Publication 175 Calibration and validation of process analysers EEMUA x EEMUA Publication 175 Calibration and validation of process analysers EEMUA 1 1. Scope: calibration and validation This Publication is a general guide to the principles and methods used for the calibration and validation (ver
36、ification) of process (on-line) analysers. It covers the principles of initial calibration and correlation, in the factory or on the plant, and initial validation. The detailed guidance is focussed on continued (routine) validation (in-service checking) and calibration adjustment (only if required,
37、based on appropriate validation results) including the use of control charts. The content and presentation of this Publication recognises that the effort available for routine validation and calibration adjustment is limited. The simplified approaches recommended are therefore intended to require on
38、ly a reasonable expenditure of effort in order to maintain assurance that the initial calibration and correlation of the analyser to the reference method is still valid. In some cases the simplified approaches will produce procedures, which although adequate, may be less rigorous than could be evolv
39、ed on purely technical and theoretical grounds. The approaches should be amended to ensure that they meet the individual business needs of the manufacturing or other process to which they are applied. It should be noted that this Publication is primarily intended for use by those directly involved w
40、ith supplying, using and maintaining on-line/process analysers. These individuals are generally not specialists in quality measurement principles and the associated statistical calculations. Therefore, as far as possible, this Publication avoids the explicit use of statistical methods and favours de
41、scriptions and explanations of a more practical nature. In order to remain consistent with the theoretical approach these practical approaches are assumed to exhibit random statistical behaviour with a normal/Gaussian distribution. This Publication is not intended to cover Continuous Emissions Monit
42、oring Systems (CEMS). These systems have calibration, validation, on-going quality assurance (routine validation and calibration adjustment) and the associated statistical tools defined by other standards, e.g. BS EN 14181 - Stationary source emissions: quality assurance of automated measuring syste
43、ms(1). This Publication is not intended to cover automatic or semi-automatic instruments used mainly in laboratories to analyse discrete samples. However, the principles and techniques described can also be applied to laboratory instruments. For consistency with BS EN 60359 - Electrical and electron
44、ic measurement equipment expression of performance(2), this Publication treats the definitions of validation and verification as equivalent. EEMUA Publication 175 Calibration and validation of process analysers EEMUA 2 EEMUA Publication 175 Calibration and validation of process analysers EEMUA 3 2.
45、Definitions The following glossary aims to add further definition to the terms used in the text in the context of their use when describing activities associated with process analysers. The definitions are grouped by a usage category (additional external references are shown in brackets). Similar wo
46、rds or phrases are listed with a common description. 2.1. Equipment Analyser / on-line analyser / process analyser / process stream analyser / quality analysers / Quality Measurement Instruments (QMI). An instrument used to measure continuously and automatically a particular property or chemical com
47、position of a process stream. The input to the instrument is a sample or portion of the process stream, either introduced continuously or at specific points in the analyser operating cycle. The sample is conditioned in a particular way, for example; flow, pressure, temperature or cleanliness. The ou
48、tput is a transmitted result or reading displayed on suitable read out instrumentation such as a control system screen or other operator interface. The analyser is normally located permanently within the process area. Analyser system / total analyser system (ASTM D3764(3). The complete instrumentati
49、on system, including sample extraction and return, sample conditioning, analytical measurement, and result display/readout/operator interface. The analyser system typically comprises the sampling system, the analyser, the read-out instrumentation and all associated piping, wiring and data communications and operator displays. Analyser time constant. For a continuous analyser this i
