VSL – Staying in control - efmws.eu · AGA 10 GPA 2145 SGERG88 GPA 2172 base conditions ISO 6976 velocity of sound GERG2004 ... 1,01325 standard pressure 0 -5,70E+03 -5,19E+03 0,00

VSL – Staying in control

“an introduction in uncertainty”

Remco van den BergRens van den Brink

Content

- Introduction: VSL, Remco, Rens

- Requirements of flow metering

- Mathematics in uncertainty calculation

- Real life applications

- Unraveling the complexity of the system

- Meeting the requirements

VSL

VSL tasks and services

Research and Development

- Development/maintenance national measurement standards

- International traceability assurance

- Research in metrology and measurement systems

Calibration and Reference materials

- Calibrations

- Reference materials

- Proficiency testing, inter-laboratory comparisons

Customized Applied Metrology

- Consultancy

- R&D Contracts

- International projects, Training

The calibration and test labs of VSL

Time and frequency

ElectricityTemperature and humidity

Length

Gas flow

Viscosity

OpticsChemistry Mass

Ionisingradiation

PressureLiquid flow and volume

Remco van den Berg

- Metrologist Flow & Volume

- 15 year in flow measurements

- Calibration of liquid, gas flow and volume meters

- Validation and certification of calibration facilities and metering systems and custody transfer

- Lab -> field ->Customized Metrology ->lab and field

- 45 years, married and two children

- Hobbies: running and

Rens van den Brink

- Metrologist at VSL Flow

- 18 years in flow measurements

- Calibration of gas flow meters (high and low pressure)

- Validation and evaluation of calibration facilities

- Maintenance and development of primary flow standards

- Training and consultancy

General requirements of metering and goals to achieve

- Conformity- To written standards, legal regulations

- Traceability- To (inter-)national measurement standards and

references

- Accuracy- Reduce systematic measurement error

- Precision- Achieve lowest possible

uncertainty

Conformity

- Select the applicable standards

- Agree on the standards with your supplier orcustomer

- Get to know the applicable standards

Written standards (1)

Organisations involved in writing standards (for natural

gas measurement):

API American Petroleum institute

AGA American Gas Association

GPA Gas Producers Association

ANSI American National Standards Institute

ASTM American Society of Testing Material

BS British Standard

EN European Norm

IP Institute of Petroleum (UK)

ISO International Organization for Standardization

OIML International Organization for Legal Metrology

10

GFM2011 Module 4B 11

gas meter standard

turbine meter EN 12261

ISO 9951

AGA 7

ultrasonic meter ISO 17089-1:2010

BS 7965

AGA 9

(domestic) EN 14236

orifice plate ISO 5167-2

AGA 3

coriolis meter AGA 11

ISO 10790

gas metering station EN 1776

uncertainty GUM

GUM suppl. 1

EA 4/02

ISO 5168

Written standards (2)

GFM2011 Module 4B 12

gas property standard gas property standard

density GERG2004 calorific value GERG2004

ISO 6976 ISO 6976

AGA 8 AGA 5

AGA 10 GPA 2145

SGERG88 GPA 2172

base conditions ISO 6976 velocity of sound GERG2004

relative density ISO 6976 AGA 10

base conditions AGA 5 viscosity

compressibility GERG2004 specific heat GERG2004

SGERG88 AGA 10

MGERG composition ISO 6974

AGA8 ISO 10723

AGA 10 ASTM 1945

AGA NX19 gas sampling ISO 10715

base conditions ISO 6976

AGA 5

Written standards (3)

Traceability

Proof it: Build the piramid:

Traceability

BIPM

VSL

Manufacturers

End users

Metrological traceability requires an established calibration hierarchy

VSL and/or

calibration

laboratoriesSI

SI

National/ Primary

Standards

Secondary / Working Standards

Instruments in the field

Calib

ratio

n h

iera

rchy

sm

all to

hig

he

r un

ce

rtain

ty

calibration hierarchy

VIM (1)

VIM (2)

VIM (3)

VIM (4)

Accuracy and Precision!

Uncertainty: The recipe

1. Describe the measurement set-up

2. Determine the mathematical modelgive the relation between all input quantities and the measurement results (output quantity)

3. Determine for each input quantity:a. The value and its uncertainty

b. The distribution function and the standard uncertainty

c. How sensitive is the measurement result for a variation in this input quantity

d. The uncertainty contribution in the measurement result

Fill in the uncertainty table

4. Determine and present the result

Simpel Mathematical Model1-Measurement setup

2-Mathematical model

FT Energy

Energy (MJ) = Quantity (m3) x Quality (MJ/m3)

Adding uncertainties2-mathematical rules

Un

cert

ain

ty

System Configuration

Component A Component B Utotaal

Contribution of uncertainty sources

With: Uflow = 0.3% rel.

UH = 0.5% rel. Uenergy = 0.58% rel.

With: Uflow = 0.2% rel.

UH = 0.5% rel. Uenergy = 0.54% rel.

With: Uflow = 0.3% rel.

UH = 0.2% rel. Uenergy = 0.36% rel.

The GUM3-Uncertainty

Type A and Type B3a-Uncertainty

Correlated and uncorrelated uncertaintysources3a-Uncertainty

Examples:

- Pressure and Z (compressibility)

- Fluctuating Flow (Reference and MUT)

- …

Distributions3b-Uncertainty

Sensitivity coefficient3c-Uncertainty

- = →

- = 1, 2 →

,

- Sens. Coeff. ci: Analytical, Numerical

Analytical examples

- = .

- =

- = ·

- = · 1+ · 2- = 2

- = √

- =

- =

Numerical examples

Math. model: Math. model:

y=x^2 y=exp(x) * sqrt(x) + (5 + x^2) / (5 + sin(x))

x 10.000 x 2.0000

dx 0.001 dx 0.0002

y (x) 100.00 y (x) 11.9727

y (x+dx) 100.02 y (x+dx) 11.9755

dy 0.02 dy 0.0028

Numerical: Numerical:

dy/dx 20.001 dy/dx 13.85 check by changing dx

Analytical: Analytical:

dy/dx = 2x 20.000 dy/dx = ???

Contribution3d-Uncertainty

- = ·

- ~

- ~

- ~!"!#

Uncertainty Table

Quantity Xi

Estimate xi Uncertaint

y

Probability distribution and k-factor

Standard deviation

u(xi)

Sensitivity coeffiicent

ci

[E/xi]

Uncertainty contribution

ui(∆E ) [E]

Uncertainty contribution

ui(∆E/∆E) [%]

∆Vn [m3] 6876 20.6 Normal

K=2 10.3 3.835E+07 7.91E+07 0.30

Hs [J.m-3] 3.835E+07 76700 Normal

K=2 38350

6.876E+03

5.27E+08 0.20

∆E [J] 2.637E+11

Expanded Uncertainty

(k=2): 1.55E+09 0.36

MODEL: Energy (J) = Quantity (m3) x Quality (J/m3)

Energy measurement systems (1)

GFM2011 Module 4B 33

Energy measurement systems (2)

GFM2011 Module 4B 34

Pressure transmitter

Relative density transducer

Gas density transducer

Gas chromatograph

Temperature transmitter

Pt 100 sensor

Energy measurement systems (3)

GFM2011 Module 4B 35

Hydrocarbon dew point

analyser

Flow Computer

Turbine meter

Orifice plate

Ultrasonic meter

GFM2011 Module 4B 36

Adjustable temperature

Adjustable pressure

Reference sensors/

read out

Energy measurement systems (4)

CRMCertified

Reference

Material

VSL equipment:

Real Life Application in Jordan

USM

PTTT

GC

Desired uncertainty = 0.5 %

Real Life App: Mathematical Model

- Real Gas

- Determination of measured Volume

$ =% ·& &%

·' '%· $% ·

1

1 + ()

- Determination of measured Energy

* =% ·& &%

·' '%· $% ·

1

1 + ()· +,

KZ

V

T

P

i

i

i

i=⋅

Real Life App: Uncertainty Budget

quantity estimate source of estimate uncertainty sensitivity uncertainty uncertainty

coefficient contribution contribution

X i x i U(x i ) c i U i ( ∆ V) U i ( ∆ V/ ∆ V)

(2s) [V/x i ] [V] [%]

P m [bar] 31 uncertainty calibration 0,0485 1,86E+02 9,04 0,16

P n [bar] 1,01325 standard pressure 0 -5,70E+03 0,00 0,00

T n [K] 288,15 standard temperature 0 2,01E+01 0,00 0,00

T m [K] 292,15 uncertainty calibration 0,20 -1,98E+01 3,96 0,07

V a [m3 ] 1,808E+02

totalisation of the flow

computer0 3,20E+01 0,00 0,00

e m [-] 0,0050 total uncertainty calibration 0,20% -5,75E+03 11,50 0,20

V n [m 3 ] 5,78E+03expanded uncertainty (k=2)

on volume measurement1,52E+01 0,26

quantity estimate source of estimate uncertainty sensitivity uncertainty uncertaintycoefficient contribution contribution

X i x i U(x i ) c i U i ( ∆ E) U i ( ∆ E/ ∆ E)[E/x i ] [E] [%]

H s [J.m-3

] 3,86E+07 uncertainty in calorific value 5,80E+04 5,78E+03 3,349E+08 0,15

E [J] 2,23E+11expanded uncertainty (k=2)

on energy measurement3,35E+08 0,30

Whatch out for the iceberg!

The visible:

Calibration uncertainties

The unvisible:

?

AGA8

Accuracy of Equation Of State (EOS)

- The mathematical model:

s

m

m

m

n

m

n

n

mH

eV

T

T

Z

Z

P

P⋅

+∆⋅⋅⋅=∆

1

1E

quantity estimate source of estimate uncertainty sensitivity sensitivity uncertainty uncertainty

coefficient coefficient contribution contribution

X i x i U(x i ) c i c i U i ( ∆ V) U i ( ∆ V/ ∆ V)

(2s) [V/x i ] [V/x i ] [V] [%]

P m [bar] 31 uncertainty calibration 0,0485 1,86E+02 1,86E+02 25,57 0,44

P n [bar] 1,01325 standard pressure 0 -5,70E+03 -5,19E+03 0,00 0,00

Z n [-] 0,997839 uncertainty of AGA8 algorithm 0,10% 5,79E+03 5,79E+03 5,78 0,10

Z m [-] 0,937705 uncertainty of AGA8 algorithm 0,09% -6,16E+03 -5,57E+03 5,78 0,10

T n [K] 288,15 standard temperature 0 2,01E+01 2,01E+01 0,00 0,00

T m [K] 292,15 uncertainty calibration 0,20 -1,98E+01 -1,98E+01 9,15 0,16

V a [m3 ] 1,808E+02

totalisation of the flow

computer0 3,20E+01 3,20E+01 0,00 0,00

e m [-] 0,0050 total uncertainty calibration 0,20% -5,75E+03 -5,23E+03 18,19 0,31

V n [m 3 ] 5,78E+03expanded uncertainty (k=2)

on volume measurement3,38E+01 0,58

quantity estimate source of estimate uncertainty sensitivity uncertainty uncertainty

coefficient contribution contribution

X i x i U(x i ) c i U i ( ∆ E) U i ( ∆ E/ ∆ E)

[E/x i ] [E] [%]

H s [J.m-3

] 3,86E+07 uncertainty in calorific value 5,80E+04 5,78E+03 3,349E+08 0,15

E [J] 2,23E+11expanded uncertainty (k=2)

on energy measurement3,35E+08 0,60

Add the Accuracy of EOS

Installation Effects

Pressure

Piping (flow profile)

Temperature

Etc…….

Vibrations

EMC

Fluid properties

- On Temperature Measurement

- On Pressure Measurement

- On Flow Meter

- On GC

Installation Effects

Inventarisation of the different

uncertainty sourcesIn practice, there are many possible sources of uncertainty in a measurement, including:

• a) incomplete definition of the measurand;

• b) imperfect reaIization of the definition of the measurand;

• c) nonrepresentative sampling — the sample measured may not represent the defined measurand;

• d) inadequate knowledge of the effects of environmental conditions on the measurement or imperfect

• measurement of environmental conditions;

• e) personal bias in reading analogue instruments;

• f) finite instrument resolution or discrimination threshold;

• g) inexact values of measurement standards and reference materials;

• h) inexact values of constants and other parameters obtained from external sources and used in the

• data-reduction algorithm;

• i) approximations and assumptions incorporated in the measurement method and procedure;

• j) variations in repeated observations of the measurand under apparently identical conditions.

GFM2011 Module 4B 45

Flip over

Give me your thoughts

GFM2011 Module 4B 46

Ambient Conditions

2-wire or 4-wire PT100

GFM2011 Module 4B 48

Distance 50m

Resistance of Copper cable 0.5 mm2 34,5Ω/km

Cable Ressitance (R0) 3,450Ω

Initial Temp (T0) 20°C

Temp Coeff (α) 0,0039Ω/°C

Actual Temp (T) 40°C

Actual Resistance (R) 3,719Ω

dR 0,269Ω

Difference in Temperature 0,70°C

- = -. 1 + / · ' − '.

Excel sheet

Where the magic happens

GFM2011 Module 4B 49

Long term Stability

-0.02

0.00

0.02

0.04

0.06

0.08

0.10

0.12

Un

cert

ain

ty [

%]

Long Term Stability @ calibration period of 2 year

(taken from Data sheet)

Flow Computer Selection

Uncertainty <-> Costs

-

The configuration of the system

Less

Me

asu

rem

en

t u

nce

rta

inty

More Less

Co

stso

f Ow

ne

rship

More

-

Maximum Profit

-20

-15

-10

-5

0

5

10

15

0 0.5 1 1.5 2 2.5 3

Mo

ne

y [

x1

0 k

€]

Uncertainty [%]

Investering in Meetsysteem Winst OmzetInvestment in system Profit Sales

Profit = f(Uncertainty)

Optimum

Where to put your Effort (=$ / € / ₽)

38%

0%8%

9%0%

13%

0%

27%

3% 2%Pm [bar]

Pn [bar]

Zn [-]

Zm [-]

Tn [K]

Tm [K]

Va [m3]

em [-]

Ct [-]

Cp [-]

More Balanced Uncertainty

27%

0%

11%

11%0%8%0%

36%

4% 3%Pm [bar]

Pn [bar]

Zn [-]

Zm [-]

Tn [K]

Tm [K]

Va [m3]

em [-]

Ct [-]

Cp [-]

Other Sources of Uncertainty or Error (1)

Flow computer Configuration

- Correct all input and output I/O’s that influence the outcome of the measurement

- Enough correction points for the I/O’S

- Use the correct formula’s

- Base conditions throughout the software

Other Sources of Uncertainty or Error (2)

Equipment

- Suitable ranges; input and output

- Suitable measurement technique

- Right I/O connections

User Errors

- During Maintenance

- During Replacement- Do not make assumptions!

VSLPO Box 6542600 AR DelftThe Netherlands

TFEI

Erik SmitsE

Remco van den BergE

+31 15 269 15 00+31 15 261 29 [email protected]

[email protected]

[email protected]

VSL group:http://lnkd.in/Bif3Sy

VSL Fluid Flow Metrology group:http://lnkd.in/DF2zJx

Questions ?

Rens van der BrinkE [email protected]