1、Designation: D 6624 06An American National StandardStandard Practice forDetermining a Flow-Proportioned Average Property Value(FPAPV) for a Collected Batch of Process Stream MaterialUsing Stream Analyzer Data1This standard is issued under the fixed designation D 6624; the number immediately followin
2、g the 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 (e) indicates an editorial change since the last revision or reapproval.INTRODUCTIONThe determination o
3、f an average property value that is representative of a batch of petroleumproduct collected and isolated in a tank or vessel has always been a challenge. Historically, theindustry practice has been to follow the appropriate procedures prescribed in Practices D 4057,D 5842,orD 4177 to extract one sam
4、ple (or a limited few, taken from top, middle, and bottom) fromthe tank or vessel after the content is mixed by any of several means to ensure the material ishomogeneous prior to sample extraction. The extracted sample is then sent to a laboratory for analysis.Depending on the property and its criti
5、cality, the average property value can also be obtained byindependently analyzing each of the top, middle, and bottom samples and the results averaged, or, thethree tank samples are mixed and testing for the property is performed on the mixture.With the introduction of in-line blending and process s
6、tream analysis in the 1960s, the potential forreal-time delivery to a pipeline, barge, ship, or tank car compartment was envisioned.To determine the average property value that is representative of a batch of product from a blendor process stream, two approaches have been developed and implemented.
7、One depends on the use ofa composite sampler, a vessel into which a sample of the flowing process or blended product streamis introduced at a flow-rate proportional to the flow-rate of the product stream (Practice D 4177). Thissample, collected over the period of time required to generate the batch
8、quantity of product, is thenanalyzed using a primary test method in the laboratory. Multiple laboratory analyses on one or morealiquots of composite sample can be averaged to provide a more precise estimate of the property valuethan a single analysis.A second technique utilizes the results produced
9、by on-line, at-line, or in-line analyticalmeasurement systems that continually test material from the process or in-line blended stream for thedesired property as it flows to a collection tank, pipeline, or shipping compartment. To determine theaverage property value of all the material collected (o
10、r shipped) at any time during the productionprocess, a unique real time flow-proportioned averaging technique evolved. By appropriate selectionof a production time period or cycle, the average property value for the collected (or shipped) materialat any time in the production or shipment cycle is ob
11、tained by recursively calculating a flow-proportion average using all available property values from the analytical measurement system and themeasured incremental quantity of product flow associated with each cycle. The determination of thisflow-proportioned average property value is the subject of
12、this practice.1. Scope1.1 This practice covers a technique for calculating aflow-proportioned average property value (FPAPV) for a batchof in-line blended product or process stream material that iscollected over time and isolated in a storage tank or vessel,using a combination of on-line or at-line
13、measurements of theproperty and flow rates.1.2 The FPAPV methodology uses regularly collected on-line or at-line process analyzer measurements, flow, andassessment of other appropriate process measurements orvalues, to calculate a flow-proportioned average property valuein accordance with flow quant
14、ity units of material produced.1This 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 Process Stream Analyzer Systems forPetroleum and Petroleum Products.Cur
15、rent edition approved July 1, 2006. Published July 2006. Originally approvedin 2001. Last previous edition approved in 2001 as D 662401.1Copyright ASTM International, 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959, United States.1.3 When the collecting vessel contains a heel (r
16、etainedmaterial prior to receipt of the production batch), both theproperty value and quantity of the heel material can bepredetermined and factored into the calculation of the FPAPVfor the new batch.1.4 This standard does not purport to address all of thesafety concerns, if any, associated with its
17、 use. It is theresponsibility of the user of this standard to establish appro-priate safety and health practices and determine the applica-bility of regulatory limitations prior to use.2. Referenced Documents2.1 ASTM Standards:2D 3764 Practice for Validation of Process Stream AnalyzerSystemsD 4057 P
18、ractice for Manual Sampling of Petroleum andPetroleum ProductsD 4177 Practice for Automatic Sampling of Petroleum andPetroleum ProductsD 5842 Practice for Sampling and Handling of Fuels forVolatility MeasurementD 6299 Practice for Applying Statistical Quality AssuranceTechniques to Evaluate Analytic
19、al Measurement SystemPerformance3. Terminology3.1 Definitions:3.1.1 analysis cycle time, nperiod of time required toproperly obtain and analyze a representative sample of theprocess stream material.3.1.2 flow-proportioned average property value (FPAPV),naverage property value of the collected materi
20、al in the tankor vessel, calculated by using the flow-proportioned averagetechnique described in the practice of all measurements per-formed on aliquots of the material while it is flowing into thetank or vessel.3.1.2.1 DiscussionThe term property as used in thispractice can be the physical, chemica
21、l, or performance propertymeasurements as provided by on-line, at-line analyzer systems,or, can be the deviation of such measurements from a desiredvalue.3.1.2.2 DiscussionThe FPAPV can include a value con-tributed by material (commonly referred to as a tank heel)present in the collection tank or ve
22、ssel before the start ofdelivery of the current process stream material.3.1.3 fit-for-use, nproduct, system, or service that issuitable for its intended use.3.1.4 linearly mixable, adjproperty is deemed to be lin-early mixable in a mass or volume measurement unit if theproperty of the mixed material
23、 can be calculated from thequantities and properties of the materials used to produce themixture.3.1.4.1 DiscussionThe general equations describing thislinearly mixable attribute are as follows:PMIXED5A1 P11 A2 P21 A3 P31 A4 P41 .1 AN PNA11 A21 A31 A41 .1 AN(1)AMIXED5 A11 A21 A31 A41 .1 AN(2)where:A
24、N= quantity of material N,PN= property of material N,PMIXED= property of mixed material, andAMIXED= quantity of mixed material.3.1.4.2 DiscussionThe material being mixed can be fromthe same process stream over time.3.1.5 total analyzer system response time, ntime intervalbetween when a step change i
25、n property characteristic at thesample loop inlet and when the analyzer output indicates avalue c corresponding to the 99.5 % of the subsequent changein analyzer results; the total analyzer system response time isthe sum of the sample loop lag time, the same conditioningloop lag time, and the total
26、analyzer response time. D 37644. Significance and Use4.1 Contractual or local regulation, or both, permitting, theFPAPV calculated according to this practice can be used torepresent the average property of the quantity of materialcollected.4.2 Due to the averaging and appropriate weighting ofanalysi
27、s results, the FPAPV estimate of the property for thecollected material is expected to be more representative andmore precise than an estimate based on a small number ofanalyses on a few samples.NOTE 1Theoretically speaking, the true property distribution for aninfinite number of batches with essent
28、ially identical FPAPVs is expectedto be Gaussian, centered at the FPAPV value, with a standard deviationthat is no less than the long term site precision standard deviation of theanalyzer system.4.3 If the measured property value can be used to predictanother property value through the use of an app
29、ropriatecorrelation equation, the FPAPV can also be used as a suitableprediction of that property.4.4 The most recently updated FPAPV can be used torepresent the property of the material currently accumulated inthe tank or vessel for process control or material dispositiondecisions, or both.5. Gener
30、al Requirements5.1 The analytical and flow measurement instrumentationsystems shall be installed in compliance with the principles setforth in API TP-550.35.2 The property being measured shall be linearly mixablewithin the range of the property measurements used to calcu-late FPAPV, and with respect
31、 to the quantity units (volume ormass).5.3 The integrity of the design, physical components andassemblies of both the analytical measurement system (inclu-sive of the sampling subsystem), and the instrumentation for2For referenced ASTM standards, visit the ASTM website, www.astm.org, orcontact ASTM
32、Customer Service at serviceastm.org. For Annual Book of ASTMStandards volume information, refer to the standards Document Summary page onthe ASTM website.3Part II Process Stream Analyzers, Manual on Installation of Refinery Instru-ments and Control Systems, available fromAmerican Petroleum Institute
33、, 1220 L St.NW, Washington, DC 200058197.D6624062flow quantity measurement shall be determined and docu-mented at the time of commissioning and at regular intervalsthereafter. Factors to be addressed shall include, but not belimited to, the following:5.3.1 The sampling system design and operation sh
34、all en-sure a sample representative of the process stream is deliveredto the applicable process stream analyzer.5.3.2 The process stream shall have dynamics such that theanalytical measurement system result used for FPAPV calcu-lation at the end of each analysis cycle time shall be represen-tative o
35、f the property of the material produced during that timeperiod.5.3.3 Analyzer functions shall be in proper condition toproduce accurate property measurement results.5.3.4 During the calculation of FPAPV, inferential or othervalidation strategies shall be in place to ensure the analyticalmeasurement
36、system results are representative of the materialin the process stream. The effectiveness of these strategies shallbe supported by data.NOTE 2Examples of these strategies can include, but are not limitedto, the following: monitoring of appropriate system parameters to ensureeach sample is representa
37、tive of the manufacturing process stream beinganalyzed; continuous comparison of an expected value of the streammeasured versus actual result; monitoring for unusually large or unusuallysmall changes of analytic data; monitoring of key analytical instrumen-tation parameters.5.3.5 Fit-for-use conditi
38、on of the flow quantity measure-ment instrumentation shall be regularly verified using suitableequipment and apparatus.5.4 The analytical measurement system shall be commis-sioned, validated, and thereafter regularly monitored using asuitably designed internal quality assurance program that meetsthe
39、 appropriate requirements and criteria of Practice D 3764and the techniques of Practice D 6299.NOTE 3It is recommended that a statistician be consulted to ensurethe proper application of Practices D 6299 and D 3764.5.5 The final FPAPV shall be calculated based on fit-for-usedata obtained from at lea
40、st 90 % of the material collected.5.5.1 For the balance of the volume (up to 10%) that has nofit-for-use analyzer data, for the purpose of calculating FPAPV,an estimate based on knowledge of the manufacturing process,or another equivalent measurement, or both, shall be used. Themaximum error associa
41、ted with this estimate that is deemedacceptable by this practice shall be no more than 2.5% of thetrue property value.5.5.2 For measurement systems that are operated in batchmode in synchronization with the manufacturing process, thequality of the initial unmeasured volume can be estimated byapplyin
42、g the analyzer result after waiting the total analyzersystem response time (Practice D 3764) from the initiation ofthe manufacturing process.5.6 For measurement systems that are operated in batchmode in synchronization with the manufacturing process (thatis, started up and shut down with the manufac
43、turing process),the statistical control status of the FPAPV measurement systemshall be validated in accordance with Practice D 6299 (orequivalent) at least once for each batch, preferably in thebeginning. The frequency of in-statistical-control validationwhen the manufacturing process is active shal
44、l commensuratewith the established stability of the measurement system.5.7 The fit-for-use condition shall be initially validated withat least 15 data points using the approach described in PracticeD 6299 for “Measurement System Bias Estimated from Mul-tiple Measurements of a Single Check Standard.”
45、 In this case,the difference between test results from aliquots of material thatare obtained and analyzed with the measurement process (testmethod) that the FPAPV is intended to represent versus theactual calculated FPAPVs are assessed. This validation shall besubsequently carried out as part of a s
46、tatistical quality assur-ance program meeting the requirement and guidelines ofPractice D 6299 for system stability.NOTE 4It is recommended that a statistician be consulted to ensurethe proper application of Practice D 6299.5.8 If the FPAPV is used to predict another property valuethrough the use of
47、 a correlation equation, the correlationequation shall be validated in accordance with PracticesD 3764, D 6299, or other appropriate statistically equivalenttechniques.6. FPAPV Calculation6.1 Determination of the flow-proportioned average prop-erty value is as follows:6.1.1 If the collecting tank or
48、 vessel contains a residualquantity of material from a previous batch (tank heel), obtain arepresentative sample of that material, test it to determine anestimate of the property value of interest, and measure orestimate the quantity of the heel material. Use the heel propertyvalue, and the heel qua
49、ntity as the starting values for FPAPVcalculation to ensure the final FPAPV value includes thecontribution from the heel material.6.1.2 For each analysis cycle, multiply the analyzer propertyvalue result times the quantity of process stream flowing duringthat cycle to determine the property 3 flow quantity value ofthe latest cycle.6.1.3 Sum the property 3 flow quantity values for all cyclescompleted through the last cycle,6.1.4 Sum the flow quantities for all cycles completedthrough the last cycle.6.1.5 Divide the sum of the property 3 flow quantity valuesby the sum
