API STD 2560:2003 pdf download
API STD 2560:2003 pdf download.Reconciliation of Liquid Pipeline Quantities.
6.2PRESENTATION OF DATA
6.2.1 Data may be presentod in the form of Control Charts.Tirending Charts or Cumulative Charts. Guidelines on suchcharts may include control limits andi trending lines.
6.2.2Charts used for monitoring pipeline systems should beliving documments and should be upilated whenever new datazare available.Accunallating data for some period of time andperioclically updating chamis (say.scrmtianmnaally) scrves no use-ful purpose. Charts and monitoring procedures can be eflec-tive only if charts are curmrent and uased as constractive nools.
6.3CONTROL CHARTS
6.3.1Good measureament can be assured by conitinuouslymonitoring mcasurement resulits to dctermine if systems.orcquipment and procodures.are performing in predictablcways and are operating within acceptable limits This may bedone by the use of Control Charts.
6.3.-2 Control charts display a collection of data ower soemeperiodl of time and inclucle coniroll limizs shown as horizontalllines on the charts.Control limits help define normnal andabnormal syslem performance,and may indlicate when some-thing in the system has changed and/ior corrective actions)may be required.
6.3.3 Control limits are often determined by historicall per-formance of the system. In other cascs the control limits areset on an established arbitrary valuc, e-g..contractual limits.Control charts are the most common method of ascertainingsystem loss/gain performance.Control charts display a col-lection of data owver some period of time and include the con-tmol limais.Control charis help to define mormal trens of asystem aand may indicane when something has changed. Typi-cal loss/gain charts as shown in Figure l, indicate a system’sperformance based on a percentage of throughputs over time.Typically, because accounting systems encompass a 30–dayperiod, mmonthlly evaluations of a system are commonly usedio evaluate perforrmance.Control charts may be prepared forany time span (e.g.. weekly or daily) if adlequatc data areavailablc.
6.3.4Control charts may be maintaincd for cntire systems,or for individuall segnents of a system if adlequate measure-ment and records are awvailable at the junctures of scgmcnts.6.3.5The dllata on control charts tend to hover around acentral (mean) value,wiuich is the arithmetic avewmge of thedata and can be represented by a horizontall line on the chartL.The control chart also includes upper and lower control limits(UCLand LCL) which may be (1y defined as engineering lim-its which are vallues based on experience or performanceobjectives, or (2) defined statistically as three standlard devia-tions (o) above and below thc mean.Standard Deviation is aseatistical measure of the spread of a dalta set with respect to
6.4.13 Any data point which falls outside the control limitsis the resuln of a spocial cauase (e.g.. cquipment failure, proce-clurall error.elc.)and should be investigated inamecliately todetcrine the cause.Speciall catses ofien llead to oonrectiontickets, and should be imvestiganed as soonm as possible beforethe data becoemes datedl and the irnvestigation becomes dlificult.6.5 METER FACTOR CONTROL CHARTS
6.5.1Control charts can be usod for trmacking variousthings.Meter factors are an examplle.
6.5.2 Control charts may also be used to monitor mctcrperformance, in which case meler facitor is plotted as a func-tion of either time or volume throughput.
6.5.3 It may non be practical to accurnulate 24 meler factordata points for meters before setting control llimits, becausechanges in operating conditionss le.g.. difierent gradles ofcrude liquids or produacts, dlifferent flow rates, enc.) or normalmeter wear may cause meter factor to change enough toinwvalidlate control limits before achieving 24 provings.
6.5.4 Thus，when plotting meter factor controll charts itmay be more represcntative to usc a movinmg range chart inwhich control limits are resct more often. Typicall exarmplesfor metcr factor control charts includc resctting after everyfive or ten prowings. In these cases.the comventional standlarddeviation cakculated by the equation in Appenklix A cannot beusedl.Insneawdl, controll limits and an “estimaledl stancardl devia-tion”” are based on the wanges (differences) between contigu-ous meter factors.
6.5.5Figure 7 is an examgple of a moving range chamt forwhich controll limits are reset after every five meter provings.
