PI Data Tuning (Exception & Compression)
PI Data Tuning (Exception & Compression)
©2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Exception & Compression Overview
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
Overview
• Provide an initial settings for Exception for filtering noises in data coming
from source.
• Tune the amount of data collected by a tag using the exception and
compression specifications
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
Data Flow
Interface Level Data Archive Level
Exception Reporting Data Compression
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
Exception Test
When a value passes this test, that value and the previous value are reported.
Why? To have a better representation of the actual tag behavior.
passes the
exception test
Temperature
Trend if previous
value is not sent
Trend if previous
value is sent
+/- ExcDev
Previous value
Snapshot value
Time
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
o Compression testing ensures that you store just enough data to accurately reproduce the original signal.
For example, in the following illustration all the events fall on the same straight line. In a simple case
like this, you do not need to store all the points on the line. If you store just two points, you can
recreate the point value for any other time.
o The compression test can essentially recreate other events through extrapolation of surrounding
events. This compression method is called swinging-door compression.
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
Last
archived
value
CompMax (time)
PI Buffer Subsystem Performance Counters
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Tag Settings per Type of Instrument
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
Quantity Unit BPCS (HMI) Display Quantity Unit BPCS (HMI) Display
Acceleration (a) Meters per sec² m/s2 Mass kilogram, tonne kg, t
Analyzers – Concentration parts per million ppm Molecular Weight grams per mol g/mol
Analyzers – Specific Gravity Specific gravity SG Molecular Weight: used on Flares Molecular Weight Mw (to match P&ID if used)
Area meters-squared m2 pH pH pH
BS&W percentage % Mega Watts, kilowatts, MW, kW,
Conductivity Microsiemens/cm μS/cm Power
Density Kilograms/m3 kg/m3 Megavars, kilovars, MVAR, kVAR
Reactive Power
Diameter Inches (in) Not applicable
meter M A
Distance, displacement, length, height, etc. (d, x, l, h, etc.) Amperage Amperes
millimeters mm
Differential pressure millimeters of H2O or Kilopascal mmH20 or kPa Voltage kiloVolts, Volts kV, V
Force Netwon N Pressure –Gauge kiloPascal kPa
Frequency Hertz Hz Pressure - Absolute kiloPascal Absolute kPaa
Pressure – Steam over 5000 kPa MegaPascal MPa
Fire & Gas – Line of Sight (LOS) Toxic Gas parts per million per meter PPM.m
Speed – Rotational Revolutions per minute rpm
Fire & Gas – Line of Sight (LOS) Combustible Gas lower explosion limit per meter LEL.m Speed - pumpjack Strokes per minute spm
Speed, velocity Meters/ sec, millimeters/ sec m/s, mm/s
Fire & Gas – Point Detectors Toxic Gas parts per million PPM Sound decible dBA
Tank Inventory rate Actual cubic meters/ hour Am3/h
Fire & Gas – Point Detectors Combustible Gas lower explosion limit LEL Temperature Celsius °C
Time (t) Hour, minute, second hr, min, sec
Flow rate – Chemical injection Litres/ hour (see note 2) l/h
Valve Opening 0%-100% (0% = Closed, 100% = Open) %
Flow rate – Mass (Saturated Steam) Tonnes/ hour (see note 1) t/h Vendor Specific
Vibration displacement Mils peak (mil pk) (as long as metric & consistent
Flow rate – Mass (Steam tonnes/hour (see note 1) t/h across the project)
Flow rate – Volume (Actual) Actual cubic meters/ hour (see notes 1 & 3) Am3/h
Vendor Specific
Flow rate – Volume (Standard) Standard cubic meters/ hour (see notes 1 & 3) Sm3/h Vibration acceleration (a) g-force (g) (as long as metric & consistent
Flow rate – Large gas flows Standard cubic meters/ hour (see note 1) e³m³/hr across the project)
Heat: Latent Heat Kilojoules/ kilogram kJ/kg Viscosity - Dynamic centistokes cS
Heating Value Net Heating Value NHV Viscosity - Kinematic centipoise cP
Heat Megajoules, Wobbe Index MJ, WI Volume – Gross, pumpable Actual cubic meters Am3
Heat: Specific Heat Kilojoules/ kilograms-kelvin kJ/(kg-K)
Volume – Standard, corrected Standard cubic meters Sm3
Level (vessels) Percentage %
Level (Tanks) Meters m (3 decimal places) Weight kilogram, gram kg, g
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
United States Customary System of Measurement Units
Quantity Unit BPCS (HMI) Display Quantity Unit BPCS (HMI) Display
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
Differential Pressure Transmitter (electronic ≤ 100 Pa) ± 1.5 PA 12M 12M ± 3 PA 3Y N/A
(Differential) Pressure Transmitter (electronic p ressure cell ≤ 5 Bar) ± 0.02 bar 12M 12M ± 0.04 bar 3Y N/A i f 20 mBa r i s re que s te d ca l i bra ti on i nte rva l l
s houl d be 6 month
(Differential) Pressure Transmitter (electronic p ressure ≤ 10 Bar) ± 0.03 bar 12M 12M ± 0.06 bar 3Y N/A a nd 20mba r i s onl y che ck a round the s e tpoi nt of
the i nte gri ty te s t.
Pressure Switch (≤ 5 Bar) ± 0.05 bar N/A 2Y ± 0.1 bar 3Y N/A
Temp erature Transmitter (RTD) ± 1ºC 12M 12M ± 2ºC 3Y N/A for ±0.2°C (for Bi ore a ctors ) pe ri odi ci ty 6M
Temp erature Transmitter (RTD) Room M onitoring ± 1ºC 12M 12M ± 2ºC 3Y N/A
Flow Transmitter (M ass Flow, M ID for fluids) ± 1 % of reading 12M 12M ± 2 % of reading 3Y N/A onl
onl yy down
down to to >> 10%
20% of me a
of pos s isbl
urie ngmera nge
as uri ng ra nge -
Flow Transmitter (Vortex, Orifice p lates with DP measurements\) ± 2 % of reading 12M 12M ± 4 % of reading 3Y N/A thi s a ppl i e s onl y i n bl a ch uti l i ti e s
± 1.5 % of reading Spa rge Ra te , 7%<Fl ow<20% of FS, ±4% of re a di ng.
if flow > 20% of FS And 20%<Fl ow<100% of FS, ±2% of re a di ng.
Flow Transmitter (Thermal M ass Flow for gases) 12M 12M ± 3 % of full scale 3Y N/A
± 3 % of reading s ta nda rd: ±0,5% Rd pl us ±0,1% FS
if flow > 7 % of FS onl y down to > 10% of me a s uri ng ra nge , mi ni mum
de ns i ty ne ce s s a ry - thi s a ppl i e s onl y i n bl a ck
Flow Transmitter (M ass Flow for gases) ± 2 % of reading 12M 12M ± 4 % of reading 3Y N/A
uti l i ti e s
Level Switch N/A 2Y 3Y N/A Us e wa te r to de te ct. On/Off che cks onl y
± 5 mm Conve rs i on mm to l i te r wi th l i ne a ri za ti on 20
Level Transmitter (guided Radar) 12M 12M ± 10 mm 3Y N/A
± 3 % of reading for level (mm) poi nts ->e rror of ± 1 % of FS (de a d vol ume , e tc...)
Weight Cell (Vessels). ≤ 50Kg ± 0.5 % of Reading 12M 12M ± 1 % of Reading 3Y N/A
Weight Cell (Vessels). ≤ 400Kg ± 0.5 % of Reading 12M 12M ± 1 % of Reading 3Y N/A
Weight Cell (Vessels). > 1000Kg ± 1 % of Reading 12M 12M ± 2 % of Reading 3Y N/A
Weight Cell (Platform Based).
≤ 150 Kg ± 0.5 % of Reading 12M 12M ± 1 % of Reading 3Y N/A
150Kg < W ≤ 1500 Kg ± 1 % of Reading 12M 12M ± 2 % of Reading 3Y N/A
> 1500Kg ± 2 % of reading 12M 12M ± 4 % of reading 3Y N/A
p H Cell (only one batch) ± 0.1 p H units every batch every batch ± 0.5 p H units 6W N/A more s te ri l i za ti ons a re done the re fore more dri ft
i s e xpe cte d
p H Cell ± 0.2 p H units 2W 2W ± 0.5 p H units 6W N/A
SOP s houl d fi x the ma xi mum numbe r of us a ge s
Redox Probe ± 20 mV 2W 2W ± 20 mV 6W N/A
Sp eed Transmitter. ± 1 rp m 2Y 2Y ± 2 rp m 3Y N/A
Pressure Relief Valves ± 2 % of set p oint N/A 2Y N/A N/A N/A
O2 Analy zer (dissolved O2) ± 2 % absolute every batch every batch ± 4 % absolute 12M N/A
Firs t s ignal (0/100 pF/cm ) calibration with
Cell density ± 10 p F/cm 12M 12M N.A. N.A. N/A
s im ulator ± 10 pF/cm
range 0-8 OD with ± 0.1 OD & range 0-1 OD with ±
UV ± 0.1 OD 12M 12M N.A. N.A. N/A
0.03 OD
TOC ± 5 % of reading 12M 12M N.A. N.A. N/A
M oisture Transmitters ± 2 % rH 12M 12M ± 4 % rH 12M N/A Me a s uri ng ra nge s ta rts onl y a t 1 µS/cm (Opte k)
± 1 µS/cm onl y pos s i bl e wi th a ra nge 1 - 50 µS/cm
± 3% of re a di ng onl y pos s i bl e a round the
Conductivity Probe 1 - 50 µS/cm (Op tek C220) ± 1 µS/cm 6M 6M N.A. 12M N/A ca l i bra ti on poi nt. I f the s e tpoi nt of fi na l ri ns e i s 5
µS/cm.
We ne e d to ca l i bra te the s e ns or wi th a buffe r of 5
Conductivity Probe 0.04 - 500 µS/cm (E&H) 3% of reading 6M 6M 6% of reading 12M N/A
Conductivity Probe 0.02- 250 mS/cm (Op tek C8086) 2% of reading 6M 6M 4% of reading 12M N/A
Conductivity Probe 2 - 500 mS/cm (E&H, CLS 54D) 2% of reading 6M 6M 4% of reading 12M N/A
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Conductivity Probe 0,5 - 2 mS/cm (E&H, CLS 54D) 4% of reading 6M 6M 6% of reading 12M N/A
Do not present this slide
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
Measurement Eng. Units HMI Display CompDevPercent CompMax CompMin Compressing ExcDevPercent ExcMax ExcMin Scan Rate
(sec)
Acceleration (a) feet per sec² ft/s2 0.2 28800 5 ON 0.1 600 0 5
ft ft
Distance, displacement, length, height, etc. Inches (in) in
0.5 28800 10 ON 0.25 600 0 10
Flow rate – Gas flow Million Standard cubic feet/ day mscfd 0.5 28800 5 ON 0.25 600 0 5
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
Measurement Eng. Units HMI Display CompDevPercent CompMax CompMin Compressing ExcDevPercent ExcMax ExcMin Scan Rate
(sec)
pH pH pH 0.1 28800 5 ON 0.05 600 0 5
Mega Watts, kilowatts, MW, kW,
Power 0.2 28800 10 ON 0.1 600 0 10
A
Current Amperes 1 28800 10 ON 0.5 600 0 10
Tank Inventory rate See Flow rate 0.5 28800 5 ON 0.25 600 0 5
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Pilot Data Tuning Results
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
CY1PIINTF01V CY1PIAR02V
PItoPI Interface
DGD ENC SKY 8 Mbps
CPU Cores: 3 2
CPU Speed [GHz]: 2.2
RAM [G B]: 128
Data
DPT KG2
CY1PIINTF02V CY1PIAR03V
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
© 2022 AVEVA Group plc and its subsidiaries. All rights reserved.
Do not present this slide
to Customers
Performance Increase
Measurements Performance Increase
Analog 38%
Digital 62%
Do not present this slide
to Customers
Performance Increase
1. Data collected from Pilot PI tags has the same timestamped values in comparison with the
Production PI tags but without the excessive repeated values scanned and archived observed
in the Production PI tags. This is due an acceptable configuration of the scan frequency and
exception reporting tag’s parameters.
2. Current Production PI tags configuration for Non-Critical PI tags are making an elevated
number of unnecessary scans to the source PI Data Archive. This is due to that current scan
frequency is set up for all PI tags at one second. For Analog tags most of them only report
change in value for a time period greater than 10 seconds or inclusive minutes. Regarding
Digital tags, changes are reported in hours or inclusive days.
3. The Scan Frequency, Exception Reporting and Compression settings used for this Pilot can be
used as guideline for tuning all Non-Critical PI tags for DPN and the rest of the rigs. However,
the values of those parameters can be improved even more and can be adjusted easily.
Questions?