AGA / API Audtng Requrements of Fscal Gas Meterng Systems and FCRI Experences R.Mascoman 1, Vjay Bhaskar 2, C.B.Suresh 3, P.N.Unnkrshnanan 4, Pambavasan 5 Dr Jacob Chandaplla 6 1 DD, 2 CM, G(I)L, 3 SRE, 4 RE, 5 JRF, 6 Drector Compressed Natural Gas Laboratory Flud Control Research Insttute, Kanjcode West, Palakkad, Kerala, Inda - 678623 E Mal. r.mascoman@fcrnda.com, Moble: 9442539140 ABSTRACT Realzaton of standardzed form of accountng of the busness leads the evolvement of formal audtng. Though started wth fnancal audts, now t encompasses, techncal actvtes, fscal measurements etc. Techncal audtng of custody quantty and qualty of gas measurement system s sgnfcantly relevant as t s drectly ted to the monetary value. Dgtal systems, whle mprovng the performance, add more complexty n the process of measurements, calbraton, mantenance etc. Sophstcaton of nstruments alone cannot guarantee the accurate relable output and assurance of the ntegrty of the measurement systems and processes s requred. Flow measurement Industry has evolved many natonal/nternatonal standards (OIML, AGA, ISO, API etc.) for prmary, secondary and tertary elements of custody transfer measurement system. Adaptaton and adherence of the standards s vtal to ensure the ntegrty of electronc gas measurement. Though the maxmum permssble tolerances specfed are 0.9% and 1% respectvely for Volume and Energy, the actual requrements are very strngent. Ths paper dscusses the functonal requrements of the fscal measurement systems, related standards. Audt objectves, detaled scope of audt, requrement of the audtors, collecton and analyss of the data, reportng audt fndngs etc are elaborated. Applcaton to typcal audt verfcaton of orfce meter, turbne meter and ultrasonc meter are presented. Method of uncertanty analyss of nstruments and practcal uncertanty estmaton are presented. Uncertantes as found and post mplementaton of recommendatons are gven. KEYWORDS Audt, Standards, mpes, Check, Balances, Data, Uncertanty 1. INTRODUCTION Installed meters n flow meterng statons need to be accurate as the mass, volume and/or energy traded are expandng and prces are ncreasng. It s mportant to sustan the requred accuracy n volume and energy. After the nstallaton /commssonng of the meterng staton, flow meter, pressure and temperature transmtter, gas chromatograph, flow computer are checked and valdated. Realzaton of standardzed form of accountng of the busness leads the evolvement of formal audtng. Though started wth fnancal audts, now t encompasses, desgn, producton, process, safety etc. Techncal audtng of fscal gas measurement system s sgnfcantly relevant as gas measurement s drectly ted to the monetary value (quantty and qualty).however the gas measurement process s very complex as the commodty s hazardous, nvsble, compressble, and 1
measurement s made nferentally. Dgtal systems, whle mprovng the performance, add more complexty n the process of measurements, calbraton, mantenance etc. Flow measurement Industry has evolved many natonal/nternatonal standards for prmary, secondary and tertary elements of custody transfer measurement system. Adaptaton and adherence of the standards s vtal to ensure the ntegrty of electronc gas measurement. Ths paper dscusses the functonal requrements of the fscal measurement systems, related standards. Audt objectves, detaled scope of audt, requrement of the audtors, collecton and analyss of the data, reportng audt fndngs etc are elaborated. Applcaton to and Typcal audt verfcaton of orfce meter, turbne meter and ultrasonc meter are presented. Method of uncertanty analyss of nstruments and practcal uncertanty estmaton for volumes/energy by of orfce meter, turbne meter and ultrasonc meter are presented. [1-2] provdes comprehensve revew of Electronc gas measurement systems and audt requrements of fscal meterng systems. 2. FISCAL METER FUNCTIONAL REQUIREMENTS The total meterng system shall comply wth the accepted standards and the natonal legslaton for volume/energy determnaton. Salent requrements are dscussed below [3]. 1. The Gas Company s responsble to make accountable measurements complant wth the applcable (legal) requrements and the latest nternatonal standards 2. The overall uncertanty on energy shall not exceed 1% [4-5].The operatonal uncertanty s wthn lmts as per standards [6-8] 3. Type B / systematc errors are actvely reduced usng agreed procedures regardless of the sgn of the devaton. 4. All parameters of the meterng system shall be fully transparent. 5. Measurements shall be traceable to nternatonal standard, ISO 17025. 6. The degree of transparency on data, qualty assurance and nstrument mantenance shall be agreed upon wth procedures for calbraton for pressure, Temperature, gas chromatograph and flow computer, ncludng tolerances and actons. 7. All nstruments are calbrated usng references, except for the flow meter. 8. The data handlng ncludng verfcatons, correctons and fnal approval of the measurng data must be documented 9. The nstruments shall be used wthn ther calbrated range. 10. The requred avalablty for the measurng nstallaton s 100%.Ths can be managed by usng redundancy n the systems (e.g.: n+1) meter runs, double GC system, No-break power).gas shall only flow f the measurng nstallaton s avalable and s functonng wthn specfcatons. 11. The equpment used shall have a formal approved status by a legal metrology department 12. Flow meter shall be mounted accordng to the standards.an upstream flow condtoner shall be used; If a flow condtoner s not used t has to be proven that there s no nstallaton effect on the flow measurement. 13. Flow meters shall be protected aganst polluton or drt. 14. Actve nterpretaton of dagnostc data of a flow meter must be used f present, for e.g. a comparson of Speed of Sound (SOS) when usng an ultrasonc meter 15. The meter run shall be suffcently thermally nsulated from ambent temperature. 16. Calbraton curve correcton shall be appled to mnmze systematc errors 2
17. Re calbraton (as found) shall be done at least every 1-5 years based on type of meter. 18. Every prmary meter shall have an ndvdual calbraton certfcate and the calbraton shall be performed complyng to the requrements of standards 19. Calbraton of the flow meters shall be performed at an nternatonally recognzed calbraton ste that s accredted accordng to ISO 17025 usng natural gas under operatonal pressure or as recommended standard. The condtons at flow calbraton shall resemble the condtons durng operaton. All parameters that may adversely affect the performance of the meter shall be consdered. All generally recognzed dfferences between the condtons at flow calbraton and condtons durng operaton shall be elmnated. If these dfferences result n a sgnfcant shft of the calbraton curve, the flow meter, the relevant upstream ppng and the flow condtoner shall be calbrated as a package. 20. B-drectonal meters are to be calbrated for both drectons wth the approprate certfcates for each drecton.the flow meters shall be adjusted such that the weghted average devaton s as close as possble to zero. After adjustment, the flow meter devatons at all the calbrated flow ponts shall be less than the maxmum permssble devaton. The errors n the flow rate range between 0.2 Qmax and Qmax shall be wthn a band of 0.3%. 21. Volumes at operatonal condtons shall be converted to volumes at agreed base condtons and shall be performed contnuously by usng lve nputs of absolute pressure, temperature and Compressblty. Compressblty (Z) shall be calculated by usng AGA/ISO standards [9-11]. 22. The flow computer, pressure transmtter and temperature transmtter shall have approval, and mounted accordng to the relevant standards. Pressure shall be measured wth an absolute pressure transmtter or wth a gauge pressure transmtter and an atmospherc pressure transmtter. Ambent nfluences such as temperature, pressure, nose, mosture, Pulsatons, sunlght etc. must be mnmzed. 23. The nputs shall be dgtal to elmnate addtonal uncertanty by transmsson technques. 24. Measured volumes shall be regstered by usng non-volatle counters. Counters shall be nstalled for volume at operatonal condtons, volume corrected for the calbraton curve, volume at base condtons and energy. 25. All fuel gas shall be measured correctly wth a correspondng flow computer. 26. A Gas chromatograph shall comply wth the followng requrements:. Type approval by authorzed body.. Mounted accordng to relevant standards.. Mnmzaton of ambent nfluences such as temperature, pressure, nose, mosture, pulsatons, sunlght etc. v. A samplng system shall ensure that gas sample s representatve to the gas n the gas staton regardng composton and condtons and tme. v. The GC should analyze all components wth an nfluence <0.01% on the calorfc value. v. The use of the pseudo component C6+ for hgher hydrocarbons s allowed only f the subdvson s constant n tme. The use of a fxed component s allowed f the concentraton s constant n tme (Carrer Gas). Ths shall be demonstrated by analyzng spot samples. 3
v. The uncertanty of the GC shall be demonstrated accordng to ISO10723. v. The GC shall be perodcally tested by analyzng a certfed natural gas, whch resembles the process gas. x. It shall be demonstrated that the GC wll functon wthn the stated uncertanty by usng a reference gas and a calbraton gas OIML [4-5] defnes maxmum permssble relatve errors (mpes), postve or negatve, for measurng systems are specfed n Table 1-3 for Accuracy Class A meters. Maxmum permssble errors on measurng volume at meterng condtons apply to type approval or Intal verfcaton where: the meter s adjusted at nomnal operatng condtons; 3. STANDARDS OIML/ISO/API/AGA documents detals manufacturng, nstallaton, operaton, and calculaton gudelnes used throughout the natural gas ndustry. OIML [4-5] Recommendaton apples to gas meters based on any measurement technology or 4. AUDIT AND RESOURCES Wth the ntroducton of electroncs n gas measurement, new challenges are encountered n audtng. The benefts of electronc gas measurement audts have been proven to justfy the costs and economcs. The two most common uses of redundancy n gas measurement are balances and check meters. Redundancy n meterng s an extremely effectve way to focus audt resources. Check meters measure the same gas as servce meter. Check meter volumes and all other data can be drectly compared to the audt staton, ncreasng the probablty of detectng errors and ncreasng the lkelhood of settlement of a dscrepancy. Acheck meter may be of prncple that s used to measure the quantty of gas that has passed through the meter at operatng condtons. It also apples to gas meters ntended to measure quanttes of gaseous fuels or other gases. Bult-n correcton devces and devces for nternal temperature compensaton are ncluded. AGA reports [6-9] contan equatons establshng the mathematcal bass for the converson process. As the reports n modfed forms have been adopted by Natonal Legal Metrology departments, complance to the reports has become mandatory for any organzaton dealng wth natural gas. API [12] covers the mnmum specfcatons for gas measurement systems of hydrocarbon and other related gases and attempts to mnmze the uncertantes assocated wth electronc gas measurement. Varous components, the algorthms necessary for adequate calculaton, specfcatons for equpment nstallaton and commssonng, and data securty requrements are dscussed. The standard also descrbes the mnmum necessary data to be made avalable n order to allow for a thorough audt of the system. Audt and Reportng Requrements are hghlghted. same or dfferent operatng prncple. They are nvaluable when falures occur at crtcal statons. Besdes falures, some types of measurement errors cannot be compensated mathematcally, such as damaged or reversed orfce plates. Check meters allow for accurate replacement of mssng or erroneous measurements and can carry tremendous weght n audt adjustment negotatons. For effectveness and relablty of the check measurement, check statons should follow the same standards/ gudelnes as appled to the audt staton. Balances or reconclaton compare volumes, energy and / or mass suppled nto a system aganst delvered from the system. The measurements gong nto the system should match the measurements gong out wthn an 4
uncertanty tolerance. Dfferences outsde the tolerance helps to solate measurement problems and apply audt resources. A good networked enterprse measurement system wll calculate the balance n dfferences automatcally. In any balancng process, care must be taken to account for all gas. Ths also opens the scope of the audt and possble error ponts to the other measurng statons ncluded n the balance. Key Process Indcators (KPIs) are another means of focusng audt and gas measurements that can be drectly assocated to those key process ndcators. Key Rsk Indcators (KRIs) can be developed whch help when and where to expand audts. An effectve audt program also nclude random audts. The cost effectveness of these random audts s enhanced wth a rsk-based approach by varyng frequency, scope and depth of evdence requred. Revenue recovered s not the only justfcaton for the cost of an audt. The potental revenue not lost must also be consdered, whch s hard to quantfy. 5. AUDIT OBJECTIVE AND SCOPE The objectve makes clear what the purpose of the audt s. The scope of the audt s the range of actvtes and the perods of records that are to be subjected to examnaton. The objectve helps to determne the scope of the audt and the best approach to take n conductng the audt whether nternal, offce, external etc. Scope determnes the breadth and depth of the audt and defnes what materal, equpment, nformaton, processes, tme perods and busness departments / enttes wll be examned. 6. AUDITS AND REVIEWS Audts provde a reasonable bass for expressng an opnon regardng the gas measurement process. Revews do not provde the same bass. In the gas measurement, revew s just an audt wth a smaller scope. Vews are also offce audts or admnstratve audts. These revews can be valuable, and less costly than a full scoped audt. Revews can be wtnessng or feld audtng to confrm that the data from the electronc gas measurement matches the condtons, settngs and szes actually n use. Audt may also nclude any data or measurement from the prmary devces to the corrected quantty transacton records, ncludng how that data s normally derved, and any supportng documentaton. Audts pull n data from regulatons, tarffs, contracts, as well as standards and the documents. Flow devce, transducers / transmtters, and the flow calculatons, prmary, secondary, tertary and auxlary devces are examned and tested, wth supportng documentaton. 7. AUDITOR If the audt s an offce revew of data, audtor wth a general knowledge of gas measurement and some gudance may be able to verfy data appled to fnal quantty transacton records. Feld audts requre more expertse. Feld audtors have to understand how nstruments functon and nteract wth each other and the process. They requre knowledge of the lmtatons of varous devces and the ablty to recognze when those lmtatons are beng exceeded. Sklls requred may nclude audtng, engneerng / techncal, and busness management consultng sklls. A team of audtors may also be requred wth dfferent members beng well-versed n dfferent aspects of the audt. The audtor should be aware of the prevalng regulatons, tarffs, ndustry standards and common ndustry practces. Understandng of the hardware, software, and operatonal specfcatons wll also help. Audtor should be able to act ndependently and unbased wth an atttude of professonal skeptcsm, crtcal of everythng, but be wllng to suggest a compromse f a wn-wn soluton s 5
unavalable or an equtable soluton 8. AUDIT DATA COLLECTION Data requested n wrtng depends on 1. Staton Characterstc Report wth detals on staton. 2. Calbraton reports of prmary flow meters, meter tubes and feld nstruments 3. Standard sample certfable wth composton, preparatory tolerance wth computed uncertanty. 4. Quantty Volume Report provdng daly, hourly volumetrc readngs wth averaged process data. 5. Alarms Lstng daly alarm condtons such as low/hgh 6. Event Report gvng detals of all staton actvty and change made to the staton parameters. 7. Meter Calbraton Test Reports ncludes complete hard copy meter reports when tests are performed or changes are made 8. Meter Change Reports when changes are made n the meterng system. 9. Gas Composton Analyss Reports wth gas composton nformaton from spot samples, or on-lne chromatography. 10. Exstng Meterng Staton polcy and Contract Documents 11. Ppng Instrumentaton dagram of Meterng Runs/ Termnals 9. EXAMINING AUDIT DATA AND INFORMATION An audtor must look for conflcts n the terms of the regulatons, tarffs, contracts, standards, and procedures aganst what s practced. If a procedure conflcts wth a contract, the terms of the contract overrde the procedures. If a contract conflcts wth a regulaton, the regulaton wll overrde the contract or the measurement system may be requred to comply wth both. If the governng documents are slent regardng standards, the scope of the audt. Audtng requres to acqure all exstng nformaton used n the determnaton of volumes. The followng are typcal source of data to perform a complete audt. 12. Techncal specfcatons of prmary secondary and tertary nstruments 13. Meter confguraton sheet of USM and Flow Computers 14. SOS and Zero Flow Test of USM 15. Flow computer Parameter Confguraton report, System Confguraton setup 16. USM dagnostc dsplay report for normal operatng flow 17. Flow computer outputs for known process varables.e. key pad nputs for AGA 3,AGA7,AGA9,GPA 2172 /AGA8/ISO 6976 calculatons 18. Flow Computer Totalzaton Checks - Flow computer/flow transmtter outputs for known process varables.e. key pad nputs ( Computer Tme must be recorded) Intal and Fnal downloads 19. Flow computer dsplay dump of varous confguratons as applcable for flow check 20. Recent GC Calbraton Report / Gas Chromatograph calculatons 21. Calbraton Certfcate of Master Pressure, Dfferental pressure calbrator, Master Temperature Transmtter calbrator, Dead weght testers, Temperature Baths the standards should be followed. The audtor determnes f the measurement system s capable of complyng wth the terms of the governng documents. The phrases Shall and shall not ndcate tems whch are requred by the standard. Should and should not reference preferred tems whch are not necessarly requred. Normatve sectons are requred. Informatve sectons are nformatonal only and are not requred. The flow computer manufacturer s software may be requred to vew the orgnal quantty transacton records, Contract nformaton can be matched to 6
the meter confguraton log. Other components of the meter confguraton log should agree wth the specfcatons of the meter, the gas qualty, and the calculaton methods. Events should be revewed for parameter changes, transducer verfcaton, and calbraton adjustments. Alarms can sometmes shed lght on falure and overrde condtons (ncludng no flow condtons), frequent out- of-range readngs, power and communcatons problems. Meter confguraton logs are reconcled to the fnal transacton records, adjusted by events, vald edts, and gas analyses. Revew of event reports ensures proper calbratons and that changes are posted properly. Electronc volumetrc data allows for ntegraton, recalculaton, and comparson by computer. The computer wll reduce tme spent on audts by automatcally flaggng potental problems. Gas chromatography has replaced almost all physcal testng for densty and heatng value. Representatve component percentages are very mportant for the analyss to be representatve. If adjustments to prevously calculated volumes are made for errors found pror to the audt, the perod and amount of the adjustment should be confrmed. Adjustment perods whch extend beyond the scope of the audt should prompt an automatc extenson of the audt through the adjustment perod. 10. AUDIT REPORTS, FINDINGS, RECOMMENDATIONS The audt culmnates n an audt report, the man medum for communcatng the fndngs and recommendatons.. It represents comments and assertons comprsng objectve, scope, fndngs, analyss, nterpretaton, suggestons, uncertanty statements etc..interested partes may desre to revew the results of the audt. 11. AUDIT ACITIVITIES OF GAS TERMINALS 1. Checkng specfcatons and sutablty of the flow meter, accessary feld transmtters for actual measurement requrements. 2. Scrutny of recent calbraton reports of transmtters wth respect to the respectve calbrated range aganst the current operatng process condtons. 3. Study of reference standard equpment used for perodc calbraton of the transmtters for ther sutablty, calbraton traceablty and uncertantes. 4. Loop checkng between the feld transmtters and the flow computers 5. Verfcaton of calculatons wth average keypad values for the process varables as per AGA 3, AGA 7, AGA 9, AGA 8 as applcable. 6. Physcal Inspecton and measurement of straght lengths of the meterng tubes/ /Thermowell etc. and Inspecton of tappng locatons of PT and RTD. 7. Assessments of uncertantes. Includng all effects 8. Calbraton of pressure, temperature transmtters and RTD probe wth reference equpment. 9. Evaluaton of envronmental condtons for possble effects on meterng electronc, mechancal equpment. 10. Scrutny of API complance of flow computer for calculatons / reportng / storage / retenton etc. 11. Verfcaton of gas chromatograph calculatons done aganst standard software. 12. Verfcaton of conformty of the meterng system as per AGA 3, AGA 7 /AGA 8/ AGA 9 ISO 6976, GPA 2172. 12. RECOMMENDATIONS FOR IMPROVEMENT 1. Installng PT 100, Class A, 4 Wre RTD wth 4 wre connectons. 2. Szng Orfce, Turbne and Ultrasonc flow meters for current meterng requrements. 3. Increase of calbraton frequency of the transmtters to mnmze the effects of 7
ambent/stablty ssues. 4. Fxng Frequency of recalbraton of Gas Chromatograph to acheve methane repeatablty less than 0.1% 5. Operatng flow rate to be greater than 10-20 % of maxmum capacty 6. Calbraton of meter at the operatng densty / operatng pressure 7. Calbraton / checkng the RTD probe at least at the average operatng temperature perodcally usng temperature bath wth calbrated reference Pt 100 class A RTD sensor. 8. Verfcaton of flow computer calculatons and flow totalzatons usng key pad data for process parameters. 9. Perodc nspecton / cleanng algned wth routne calbraton 10. Flow meter calbraton n an ISO 17025 accredted laboratory wth mnmum uncertanty. 11. Incorporaton of flow rates / frequences / meter errors / K factors approprately n the flow computer to mprove the measurement accuracy 12. Calbraton of master equpment n accredted laboratory wth lowest possble uncertantes. 13. Usng temperature bath only as a source 14. Implementaton of Detal Characterzaton Method n flow computers 15. Insulaton of transmtter enclosure/ 13. UNCERTAINTY EVALUATION IN VOLUME AND ENERGY Table 4 gves defnton of some mportant terms related to uncertanty. 1. Express n mathematcal terms the dependence of the measurand, output quantty Y on the nput quanttes X. 2. Identty and apply all sgnfcant correctons to the nput quanttes from calbraton. 3. Lst all sources of uncertanty n the form of an uncertanty budget thermo well/rtd/pressure Impulse tubes / upstream / downstream meterng tubes to control the effect of ambent temperature varatons. 16. RTD shall be n contact wth the bottom of thermo well wth sutable thermally conductng flud n the thermo well. 17. Updatng flow computers nstantaneously on completon of analyss 18. Calbraton range for Calbrators / Temperature Bath/ Temperature Transmtter/ RTD both n source /measure modes to sut flow measurement purpose wth more number of calbraton ponts n the range. 19. Verfcaton of master lst of confguraton data/ constant parameters nputted n the flow computer. 20. Provson of shelter for the meterng skd components 21. Installaton of flow profler / thermo well / meter tubes as per recommended nstallaton. 22. Use of certfed dameters n Flow computer for flow calculatons. 23. Revew of Staton Desgns n vew of current operatng flow rate. 24. Calbraton of the pressure/dfferental pressure calbrators aganst dead weght tester n accredted laboratory wth lowest possble uncertanty. 4. Calculate the standard uncertanty u ( x ) for repeatedly measured quanttes (Type A) 5. For sngle values, e.g. a sngle measured value, a resultant value of prevous measurements, correcton values or a reference value from lterature () adopt the standard uncertanty when t s gven () calculate from unequvocal uncertanty data () evaluate based on experence (v) calculate from manufacturers data (v) data provded n calbraton or other certfcates (v) uncertanty assgned to reference data taken from handbooks. When a probablty dstrbuton s assumed for the quantty X, based on experence or theory, then expected value and the standard devaton of the 8
dstrbuton s taken as the estmate x and the assocated standard uncertanty u respectvely. ( ) x 6. For nput quanttes for whch probablty dstrbuton s known or can be assumed, calculate the standard u. x uncertanty ( ) 7. Calculate for each nput quanttyx, the contrbuton u ( y) to the uncertanty assocated wth the output estmate resultng from nput estmate x and sum ther squares to obtan the square of the standard uncertanty u ( y) of the measurand. 8. Calculate the expanded uncertanty U by multplyng the standard uncertanty u ( y) assocated wth the output estmate by a coverage factork. 9. Report the result of the measurement comprsng the estmate y of the measurand, the assocated expanded uncertanty U and the coverage factor k n the calbraton certfcate. For the measurement of flow rate, t s necessary to combne the values of a number of nput quanttes to obtan a value for the output. It s assumed that the sources of uncertanty are uncorrelated: If the functonal relatonshp between the nput quanttesx 1, X 2,.. X n and the output quantty Y n a flow measurement process s specfed as ( X X ) Y = f,,... 1 2 X, X n (1) then an estmate of Y, denoted by y, s obtaned from Equaton (2) usng nput estmates x, x,... 2... x 1 ( x x ) y = f,,... 1 2 x, x n (2) n The total uncertanty of the process may be found by calculatng and combnng the uncertanty of each of the contrbutng factors usng u N c (y) ) = 1 [ cu(x ] 2 = (3) Type B evaluaton of uncertanty s carred out by means other than the statstcal analyss of a seres of observatons. In case of flow measurement audt, only Type B uncertantes are nvolved. Type B evaluaton of uncertanty requres knowledge of the probablty dstrbuton assocated wth the error/reported uncertantes n certfcates. Rectangular probablty dstrbutons nclude Instrument drft between calbratons, Error due to lmted resoluton of an nstrument's dsplay, Manufacturer s tolerance lmts. When the upper and lower lmts of an nput quantty X are possble to estmate wthout any specfc knowledge of concentraton of x wthn the lmts, then t can be assumed that t s equally probable for X to be anywhere wthn ths nterval. The uncertanty s calculated from ( ) / 3 u x = a (4) where a s the half-range of the dstrbuton. Normal Probablty Dstrbuton ncludes calbraton certfcates quotng a confdence level of coverage factor wthn the uncertanty ( x ) u Expanded uncertanty = (5) k where k s 2 for 95% confdence level. For combnng uncertantes.t s necessary to consder the effect of each nput quantty on the fnal result. e. the senstvty of an output quantty to an nput quantty - the senstvty coeffcent. When the functonal 9
relatonshp s specfed, then the senstvty coeffcent s defned as the rate of change of the output quantty y wth respect to the nput quanttyx, and the value s obtaned by partal dfferentaton of the functon. c y = (6) x For Type B evaluaton, when lower and upper lmts are not known ν =α. Tables 5-7 gves the detals of typcal uncertanty budget for the meters. Fgs. 1-3 presents as found and achevable uncertantes after mplementatons of audt recommendatons. Uncertantes are mproved sgnfcantly on mplementaton 14. CONCLUSION Natural gas measurement s truly a complcated, specalzed task. Possblty of errors and mstakes makes audtng a necessty. Fscal Measurement s responsble for balancng the purchase and sale of Natural gas. Meterng Errors affect the balance of busness. Audtng not only becomes desrable, but necessary. Audt resources, ncludng the expertse of the audtor, can be appled wth varous objectves and scopes to capture the bulk of measurement errors and potental errors. The fnal result of ths entre audtng process s an assurance of the ntegrty of an organzaton s electronc gas measurement busness process. API 21.1 provdes the gudance necessary to ensure the ntegrty of electronc gas measurement values. API 21.1 helps the audtor to transform the complexty of electronc gas measurement nto an effectve means of evaluatng fnancal rsk and fnancal opportunty. Uncertanty assessments confrms that after mplementaton of audt recommendatons, the values mproves sgnfcantly 15. ACKNOWLEDGEMENTS Author records hs thanks and acknowledgments to cted natonal and nternatonal references/organzatons for the relevant publshed data and nformaton used whle wrtng ths paper. 16. REFERENCES 1. Keth Fry Audtng Electronc Gas Measurement as Per API Chapter 21.1,Quorum Busness Solutons, Inc. Dallas, TX 75247, USA 2. Kenneth W. Blackburn Electronc Gas Measurement Audtng Houston Flow Measurement, Inc., TX 77025,USA 3. Functonal Requrements for Fscal Meterng Systems,GTS,LGM,2010 4. OIML R 137-1&2 Edton 2012 : Gas meters 5. OIML R 140 2007 (E) : Measurng system for Gaseous Fuels 6. AGA Report No 3:2000: Orfce Meterng of natural gas and other related hydrocarbon fluds, Part 2, Specfcatons and Installaton Requrements 7. AGA Report No 7:2006: Measurement of Natural Gas by Turbne Meters\ 8. AGA Report No 9, 2007 Measurement of Gas by Multpath Ultrasonc Meters 9. AGA Report No 8:1992: Compressblty Factors of Natural Gas and other related Hydrocarbon Gases. 10. GPA 2172:2009: Calculaton Of Gross Heatng Value, Relatve Densty, Compressblty And Theoretcal Hydrocarbon Lqud Content For Natural Gas Mxtures For Custody Transfer 11. ISO 6976:1995 Natural gas -- Calculaton of calorfc values, densty, relatve densty and Wobbe ndex from composton. 12. API MPMS, Chapter 21.1 : Flow Measurement Usng Electronc 10
Table 1 mpes for measurng systems Maxmum permssble errors on determnng: Accuracy class A Converted volume, converted mass or drect mass ± 0.9 % (<0.3 %) Energy ± 1.0 % (<0.75 %) Table 2 - mpes for modules Maxmum permssble errors on: Accuracy class A Measurng volume at meterng condtons ± 0.70 % (<0.3%) Convertng nto volume at base condtons or nto mass ± 0.50 %(<0.01 %) Calorfc value measurement (only CVDD) ± 0.50 %(<0.2 %) Maxmum permssble errors on: Table 3 mpes for nstruments Accuracy class A Temperature ± 0.5 C (<0.27 C ) Pressure ± 0.2 % (<0.1%) Densty ± 0.35 %((<0.2%)) Compressblty ± 0.3 %(<0.1%) Values wthn ( ) are current practce Table 4 Uncertanty Standard Uncertanty, u ( x) Combned Standard Uncertanty, u c ( y) Expanded Uncertanty, U = ku c ( y) Parameter, assocated wth the result of a measurement, that Characterzes the dsperson of the values that could reasonable be attrbuted to the measured. For practcal purposes, the uncertanty s taken to be twce the Standard devaton of a dstrbuton. Also, the probabltes are assumed to be normally dstrbuted. Ths mples that there s a 95%probablty to reman wthn the uncertanty boundares and that values are dstrbuted symmetrcally around the expectaton value. Uncertanty of the result of measurement s the standard error expressed as a standard devaton of mean. Standard uncertanty of the result of a measurement when that result s obtaned from the values of a number of other quanttes, equal to the postve square root of a sum of terms, the terms beng the varances of these other quanttes weghted accordng to how measurement result vares wth changes n these quanttes. Quantty defnng an nterval about the result of a measurement that may be expected to encompass a large fracton of the dstrbuton of values that could reasonably be attrbuted to the measurand. The fracton may be vewed as the coverage probablty or the level of confdence of the nterval. 11
Coverage factor,k Type A Evaluaton Type B Evaluaton Numercal factor used as a multpler of the combned standard uncertanty n order to obtan an expanded uncertanty s typcally n the range 2 to 3. Method of evaluaton of uncertanty by the statstcal analyss of a seres of observatons. Method of evaluaton of uncertanty by means other than the statstcal analyss of a seres of observatons. Table 5 : Uncertanty n Orfce Meterng System (Typcal) Expanson coeffcent,γ D 0.00001652 K -1 0.0000004 K -1 Type B, Rectangular 1.7321 0.000000 K -1 1.620000 m.k 3.7412E-07 Reference temperature,t o 20.00 o C 0.50 o C Type B, Rectangular 1.7321 0.288675 K -2.676E-06 m/k -7.726E-07 Flowng temperature,t f 30.0000 o C 0.37 o C Type B, Rectangular 1.7321 0.215388 K 2.676E-06 m/k 5.7643E-07 5 Ppe dameter,d 0.305034 m 0.000008 m Type B, Normal 2 0.000004 m Reference dameter,d To 0.305 m 0.000007 m Type B, Normal 2 0.000004 m 1.0001107 3.6504E-06 Expanson coeffcent,γ D 0.000011071 K -1 0.0000004 K -1 Type B, Rectangular 1.7321 0.000000 K -1 3.050000 m.k 7.0437E-07 Reference temperature,t o 20.00 Flowng temperature,t f 30.0000 o C 0.50 o C 0.37 Uncertanty Budget for Orfce Meterng System Sl.No Source of Uncertanty Estmate Expanded Uncertanty Degree of Probablty dstrbuton Standard Uncertanty Senstvty coeffcent uncertanty freedom x x U(x) Type Dvsor u(x) c u(y) ע Value unt Value unt Value unt Value unt 1 Coeffcent of dscharge,c d 0.604264-0.002714 - Type B, Rectangular 1.7321 0.001567-4.633619 kg/s 0.007260 2 Velocty factor,ev 1.042350-0.000010 - Type B, Normal 2 0.000005-2.686170 kg/s 0.000013 Orfce dameter,d 0.162027 m 0.000008 m Type B, Normal 2 0.000004 m 1.112856 1/m 0.000004 Ppe dameter,d 0.305034 m 0.000008 m Type B, Normal 2 0.000004 m -0.591123 1/m -0.000002 Beta rato,β 0.531176-3 Expanson factor,y 0.993699-0.074831 % Type B, Rectangular 1.7321 0.000429-2.817684 kg/s 0.001210 4 Orfce dameter,d 0.162027 m 0.000008 m Type B, Normal 2 0.000004 m 34.561318 kg/s/m 0.000131 Reference dameter,d To 0.162 m 0.000007 m Type B, Normal 2 0.000004 m 1.0001652 3.6506E-06 o C Type B, Rectangular 1.7321 0.288675 K -3.377E-06 m/k -9.748E-07 o C Type B, Rectangular 1.7321 0.215388 K 3.377E-06 m/k 7.2729E-07 6 Densty 3.138239 kg/m 3 0.006067 kg/m 3 Type B, Rectangular 1.7321 0.003503 kg/m 3 0.446099 m 3 /s 0.001563 Pressure,P f 400905 Pa 87.14356 Pa Type B, Normal 2 43.572 Pa 7.828E-06 kg/m 3 /Pa 0.00034107 Molecular Mass,Mr 19.52983 kg/kmol 0.0096317 kg/kmol Type B, Rectangular 1.7321 0.005561 kg/kmol 0.161 kmole/m 3 0.00089358 Compressblty,Z f 0.989834-0.000990 - Type B, Rectangular 1.7321 0.000571 - -3.170 kg/m 3-0.0018119 Unversal gas constant,r 8314.46 J/kmol/K 0.831446 J/kmol/K Type B, Rectangular 1.7321 0.480036 J/kmol/K 0.000 kg/m 3 /(J/kmol/K) -0.0001812 Temperature,T f 303 K 0.37 K Type B, Rectangular 1.7321 0.215388 K -0.010 kg/m 3 /K -0.0022297 7 Dfferental Pressure 7500 Pa 4.1231056 Pa Type B, Normal 2 2.061553 Pa 0.000187 kg/s/pa 0.000385 8 Base Densty 0.828498 kg/m 3 0.001071 kg/m 3 Type B, Rectangular 1.7321 0.000618 kg/m 3-4.079095 sm 3 /s -0.002522 Pressure,P b 101325 Pa 0 Pa Type B, Normal 2 0.000 Pa 8.177E-06 kg/m 3 /Pa 0 Molecular Mass,Mr 19.52983 kg/kmol 0.0096317 kg/kmol Type B, Rectangular 1.7321 0.005561 kg/kmol 0.042 kmole/m 3 0.0002359 Compressblty,Z b 0.996944-0.000997 - Type B, Rectangular 1.7321 0.000576 - -0.831 kg/m 3-0.0004783 Unversal gas constant,r 8314.46 J/kmol/K 0.831446 J/kmol/K Type B, Rectangular 1.7321 0.480036 J/kmol/K 0.000 kg/m 3 /(J/kmol/K) -4.783E-05 Temperature,T b 288.15 K 0.00 K Type B, Rectangular 1.7321 0.000000 K -0.003 kg/m 3 /K 0 9 Mass Flow Rate 2.79993 kg/s 0.015 kg/s Type B, Rectangular 1.7321 0.008700 kg/s 1.207003 sm 3 /kg 0.01050147 Actual Volume Flow Rate 3211.91 m 3 /h Flow Velocty 12.21 m/s 10 Uncertanty n Mass flow rate 0.538 % 11 Base Volume Flowrate 3.37952 sm 3 /s 0.0216 sm 3 /s 12 Uncertanty n Volume flow rate 0.639 % 13 Base Volume Flowrate 12166.283 sm 3 /h 77.8 sm 3 /h Reynolds Number 717947.3 14 Uncertanty n Volume flow rate 0.639% Base Voulme 12166.283 sm 3 /h 77.7608 sm 3 /h Type B, Rectangular 1.7321 44.8952243 sm 3 /h 43.943 MJ/sm 3 1972.84161 Gross Calorfc Value 43.943 MJ/sm 3 0.0754946 MJ/sm 3 Type B, Rectangular 1.7321 0.04358681 MJ/sm 3 12166.283 sm 3 /h 530.289528 15 Uncertanty n Energy Flowrate 534625.9 MJ/h 4085.737 MJ/h 0.76% Post Implementaton of Recommendatons : RTD ( Pt 100 Class A ): 0.27 C : Pressure Transmtter 87 Pa : Dfferental Pressure Transmtter 4Pa : Flow Computer : FB 107 (Ins) : 0.24 C GC Repeatablty 0.1% 12
Table 6 : Uncertanty n Turbne Meterng System (Typcal) Uncertanty Budget for Turbne Meterng System Uncertanty Degree of Source of uncertanty Estmate Expanded uncertanty Probablty dstrbuton Standard Uncertanty Senstvty coeffcent Sl.No freedom x x U(x) Type Dvsor u(x) c u(y) ע Value unt Value unt Value unt Value unt 1 Frequency 100.31570-0.000000 Pulses/s Type B, Rectangular 1.7321 0.000000-0.000941 m 3 /pulses 0.000000 2 K Factor 1062.86000-3.720010 Pulse/m 3 Type B, Normal 2 1.860005 - -0.000089 m 6 /s/pulses -0.000165 3 Densty 4.881860 kg/m 3 0.009820 kg/m 3 Type B, Rectangular 1.7321 0.005670 kg/m 3 0.094383 m 3 /s 0.000535 Pressure,P f 680860 Pa 490.3325 Pa Type B, Normal 2 245.166 Pa 7.17E-06 kg/m 3 /Pa 0.00175788 Molecular Mass,Mr 17.21904 kg/kmol 0.010335436 kg/kmol Type B, Rectangular 1.7321 0.005967 kg/kmol 0.284 kmole/m 3 0.00169178 Compressblty,Z f 0.985275-0.000985 - Type B, Rectangular 1.7321 0.000569 - -4.955 kg/m 3-0.0028185 Unversal gas constant,r 8314.46J/kmol/K 0.831446 J/kmol/K Type B, Rectangular 1.7321 0.480036 J/kmol/K -0.001 kg/m 3 /(J/kmol/K) -0.0002819 Temperature,T f 293.150 K 0.33 K Type B, Rectangular 1.7321 0.191138 K -0.017 kg/m 3 /K -0.003183 4 Base Densty 0.728510 kg/m 3 0.000985 kg/m 3 Type B, Rectangular 1.7321 0.000569 kg/m 3-0.868175 sm 6 /s/kg -0.000494 Pressure,P b 101325 Pa 0 Pa Type B, Normal 2 0.000 Pa 7.19E-06 kg/m 3 /Pa 0 Molecular Mass,Mr 17.21904 kg/kmol 0.010335 kg/kmol Type B, Rectangular 1.7321 0.005967 kg/kmol 0.042 kmole/m 3 0.00025246 Compressblty,Z b 0.997687-0.000998 - Type B, Rectangular 1.7321 0.000576 - -0.730 kg/m 3-0.0004206 Unversal gas constant,r 8314.46J/kmol/K 0.831446 J/kmol/K Type B, Rectangular 1.7321 0.480036 J/kmol/K 0.000 kg/m 3 /(J/kmol/K) -4.206E-05 Temperature,T b 288.71 K 0.00 K Type B, Rectangular 1.7321 0.000000 K -0.003 kg/m 3 /K 0 5 Actual Volume Flow rate 0.094383 m 3 /s 0.000330 m 3 /s Type B, Rectangular 1.7321 0.000191 m 3 /s 4.882 kg/m 3 0.00093105 6 Mass Flow Rate 0.4608 kg/s 0.002148 kg/s Type B, Rectangular 1.7321 0.001240 kg/s 1.3726653 sm 3 /kg 0.0017021 7 Uncertanty n Mass flow rate 0.466125462 % 8 Base Volume Flowrate 0.6325 sm 3 /s 0.004 sm 3 /s 9 Uncertanty n Volume flow rate 0.560 % 10 Base Volume Flowrate 2276.907 sm 3 /h 12.760 sm 3 /h Type B, Rectangular 1.7321 7.366980 sm 3 /h 39.969 MJ/sm 3 294.451559 Uncertanty n Base Volume 11 flow rate 0.560 % 12 Gross Calorfc Value 39.969 MJ/sm 3 0.075 MJ/sm 3 Type B, Rectangular 1.7321 0.043380 MJ/sm 3 2276.907 sm 3/ /h 98.7731713 13 Uncertanty n Energy Flow rate 91005.932 MJ/h 621.153 MJ/h 14 0.683 % Post Implementatons of Recommendatons : Uncertanty n K Factor : 0.35 % ; GC Repeatablty : 0.1%: Uncertanty n Pressure :490 Pa ; Uncertanty n Temperature (TT/RTD ) : 0.27 C Table 7: Uncertanty n Ultrasonc Meterng System (Typcal) Sl.No Source of uncertanty Estmate Uncertanty Budget for Ultrasonc Meterng System Expanded uncertanty Probablty dstrbuton Standard Uncertanty Senstvty coeffcent Uncertanty contrbuton x x U(x) Type Dvsor u(x) c u(y)=c*u(x) ע Value unt Value unt Value unt Value unt Value 1 Velocty 4.00000 m/s 0.012400 m/s Type B,Normal 2 0.006200 m/s 6169.989778 sm 3 /h/m/s 38.253937 2 Meter dameter,d 0.188895 m 0.000070 m Type B, Normal 2 0.000035 m 261309.147376 sm 3 /h/m 9.145820 3 Expanson coeffcent,γd 0.00001512 K -1 0.0000004 K -1 Type B, Rectangular 1.7321 0.000000 K -1 15.002268 K 3.46463E-06 4 Reference temperature,to 20.00 C 0.25 K Type B, Rectangular 1.7321 0.144338 K -0.000045 /K -6.54814E-06 5 Flowng temperature,tf 25 C 0.27 K Type B, Rectangular 1.7321 0.155834 K 0.000045 /K 7.06971E-06 6 Temperature Correcton Factor 1.0002268 0.00002 - Type B, Rectangular 1.7321 0.000012-24674.362545 sm 3 /h 0.291751681 7 Calbraton Reference Pressure 101325 Pa 101.33 Pa Type B,Normal 2 50.662500 Pa -4.93659E-11 /Pa -2.501E-09 8 Meter Pressure,Pf 5788743.991 Pa 3553.47 Pa Type B, Normal 2 1776.737 Pa 4.93659E-11 /Pa 8.77102E-08 9 Pressure Correcton Factor 0.0002808 0.0000002 - Type B, Rectangular 1.7321 0.000-24673.031801 sm 3 /h 0.00249987603 10 Lne Densty 41.865764 kg/m 3 0.076456 Kg/m 3 Type B, Rectangular 1.7321 0.044142 kg/m 3 589.502173 sm 6 /h/kg 26.021760 11 Meter Pressure,Pf 5788743.991 Pa 3553.47 Pa Type B, Normal 2 1776.737 Pa 7.23227E-06 kg/m 3 /Pa 0.01284984 12 Molecular Mass,Mr 16.19314 kg/kmol 0.01010139 kg/kmol Type B, Rectangular 1.7321 0.005832 kg/kmol 2.585 kmole/m 3 0.015078169 13 Compressblty,Zf 0.903214-0.000903 - Type B, Rectangular 1.7321 0.000521 - -46.352 kg/m 3-0.02417121 14 Unversal gas constant,r 8314.4 J/kmol/K 0.831440 J/kmol/K Type B, Rectangular 1.7321 0.480032 J/kmol/K -0.005 kg/m3/(j/kmol/k) -0.002417121 15 Temperature,Tf 298.15 K 0.27 K Type B, Rectangular 1.7321 0.155834 K -0.140 kg/m 3 /K -0.021882015 16 Base Densty 0.684900 kg/m 3 0.000935 Kg/m 3 Type B, Rectangular 1.7321 0.000540 kg/m 3-36034.410703 sm 6 /h/kg -19.461800 17 Pressure,Pf 101325 Pa 0 Pa Type B, Normal 2 0.000 Pa 6.75944E-06 kg/m 3 /Pa 0 18 Molecular Mass,Mr 16.19314 kg/kmol 0.01010139 kg/kmol Type B, Rectangular 1.7321 0.005832 kg/kmol 0.042 kmole/m 3 0.00024667 19 Compressblty,Zf 0.997994-0.000998 - Type B, Rectangular 1.7321 0.000576 - -0.686 kg/m 3-0.000395427 20 Unversal gas constant,r 8314.4 J/kmol/K 0.831440 J/kmol/K Type B, Rectangular 1.7321 0.480032 J/kmol/K 0.000 kg/m3/(j/kmol/k) -3.95427E-05 21 Temperature,Tf 288.71 K 0.00 K Type B, Rectangular 1.7321 0.000000 K -0.002 kg/m 3 /K 0 22 Volume Flowrate 403.750 23 Base Volume Flow Rate 24679.95911 sm 3 /h 102.038982 sm 3 /h Type B, Normal 2 sm 3 /h 51.019491 24 0.41345 % 25 Base Volume flow rate 24679.96 102.039 sm 3 /h Type B, Rectangular 1.7321 58.912234 sm 3 /h 38.007 MJ/sm 3 2239.083153 26 Gross Calorfc Value 38.007 MJ/sm 3 0.075 MJ/sm 3 Type B, Rectangular 1.7321 0.043200 MJ/sm 3 24679.959 sm 3/ /h 1066.171233 27 Uncertanty n Energy Flow rate 938013.7 MJ/h 4959.925 MJ/h 0.529 % RTD ( Pt 100 Class A ) : Pressure Transmtter 3554 Pa : Temperature Transmtter( Ins) : GC Repeatablty 0.1%: USM Expanson corrrecton wth 1.01325 bar and 20 o C and operatng pressure and temperature to be done n FC. Pressure coeffcent shall be calculated for operatng condtons and nputted n FC. Calbraton corrrecton for velocty done n USM (assumed) Degree of freedom 13
Uncertanty, % 7.5 7 6.5 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 Energy (AF) Volume (AF) Energy (R) Volume (R) Uncertantes n Orfce meters AF : As Found R : Recommended 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 Termnal Nos Fg.1 Uncertanty, % 1 0.95 0.9 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 0.45 0.4 0.35 0.3 Uncertantes n Ultrasonc meters Energy (AF) Volume (AF) Energy (R) Volume (R) AF : As Found R : Recommended 0.25 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 Termnal Nos Fg.2 Uncertantes of Turbne meters Uncertanty, % 3.5 3.25 3 2.75 2.5 2.25 2 1.75 1.5 1.25 1 0.75 0.5 0.25 Energy (AF) Volume (AF) Energy(R) Volume (R) 0 1 2 3 4 5 6 7 8 9 10 11 12 Termnal Nos Fg.3 14