0% found this document useful (0 votes)
126 views28 pages

Curtis Experience Using The PI System For Full Featured 1

The document discusses Portland General Electric's experience using the PI System for vibration monitoring. It describes PGE's initial fragmented vibration monitoring systems across different plants and machines. It outlines problems with the previous system's lack of online condition monitoring, remote access, integrated data, and ability to easily share asset health information. The document then details how implementing the PI System and Bently Nevada's SETPOINT solution enabled online condition monitoring, remote access, integrated data in a single system, easy data sharing, and building asset health dashboards.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
126 views28 pages

Curtis Experience Using The PI System For Full Featured 1

The document discusses Portland General Electric's experience using the PI System for vibration monitoring. It describes PGE's initial fragmented vibration monitoring systems across different plants and machines. It outlines problems with the previous system's lack of online condition monitoring, remote access, integrated data, and ability to easily share asset health information. The document then details how implementing the PI System and Bently Nevada's SETPOINT solution enabled online condition monitoring, remote access, integrated data in a single system, easy data sharing, and building asset health dashboards.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 28

Experience using the PI System for

Vibration Monitoring at PGE


Austin Curtis, PGE | Steve Sabin, B&K Vibro

#PIWorld ©2019 OSIsoft, LLC 1


About Portland General Electric
• Serves ~885 thousand customers in Portland, Salem, and
surrounding Oregon communities (51 cities total)
• 2900 employees
• Five Natural Gas-Fired Plants [CATEGORY
[CATEGORY
NAME]
Plant Capacity NAME]
[PERCENTA
Beaver 509 MW [PERCENTA
GE]
GE]
Carty 437 MW
Coyote Springs 249 MW
[CATEGORY
Port Westward 1 411 MW NAME] [CATEGO
[PERCENTA NAME
Port Westward 2 225 MW [CATEGORY
NAME] GE]
[PERCEN
[PERCENTA #PIWorld ©2019 OSIsoft, LLC GE]
2
GE]
About PGE’s Beaver Plant
• Operating since 1974
• 6-on-1 combined cycle
• Units 1-6: 6 x GE 7B
combustion turbines
• Unit 7: 1 x 155MW
GE steam turbine
• Unit 8: 1 x 25 Alstom
simple-cycle Beaver Plant
Clatskanie, Oregon
combustion turbine

#PIWorld ©2019 OSIsoft, LLC 3


How our journey began…
July 17, 2015

Hi, I ran across your PI usergroup


presentation and I'm interested in
more information on your CBM.
Portland General Electric operates
three combined cycle, ~400 wind
towers, a sub-critical coal plant, and
two large hydro projects. We're
interested in centralizing our
monitoring efforts for critical
equipment and adding balance of
plant equipment to this effort.

Thanks,
Ken Kubo

#PIWorld ©2019 OSIsoft, LLC 4


Vibration Monitoring Systems (As-Found)
Unit Machine Size Sensors MPS CMS

1 GE Frame 7B* 60MW Velocity GE Speedtronic None


2 GE Frame 7B* 60MW Proximity GE Speedtronic None
3 GE Frame 7B* 60MW Velocity GE Speedtronic None
4 GE Frame 7B* 60MW Velocity GE Speedtronic None
5 GE Frame 7B* 60MW Velocity GE Speedtronic None
6 GE Frame 7B* 60MW Proximity BN3500 None
7 GE steam turbine 155MW Proximity BN3500 None
8 Alstom GT10B2* 25MW Proximity BN3500 None
Typical as-found monitoring
* Combustion turbines system (Bently Nevada 3500
Series on Beaver Unit 8)

#PIWorld ©2019 OSIsoft, LLC 5


Velocity Vibration Measurements

Velocity
(in/sec 0-pk)

#PIWorld ©2019 OSIsoft, LLC 6


Proximity Vibration Measurements

Y-probe X-probe

Displacement
(mils pk-pk)

#PIWorld ©2019 OSIsoft, LLC 7


Problems

1. No online ConMon = no data when needed


• Reactive – not there when problems occur
• Gather data “after the fact” (problem may no longer be
present); scene of the crime may be devoid of evidence.
• Often MISS critical data – no means to forecast and model

#PIWorld ©2019 OSIsoft, LLC 8


Problems

2. No remote access
• Specialist flies to site to collect data manually
• Requires moving people, not data; involves plane tickets and
5-hour travel times
• SLOW

#PIWorld ©2019 OSIsoft, LLC 9


Problems
3. Siloed data
• Incumbent provider’s CM
software entails a separate
“silo” just for vibration data
• Integration effort to combine
and determine cause/effect VIBRATION PROCESS
• Additional IT burden DATA DATA
(2 systems)
• Not in PI = Not accessible to
other analytic tools such as
APR

#PIWorld ©2019 OSIsoft, LLC 10


Problems

4. “Closed” data
• Special viewer software and licensing required
• Hard to share data when/where needed
• FTP sites required to share large files
• Impedes collaboration to solve problems

#PIWorld ©2019 OSIsoft, LLC 11


Problems

5. Can’t easily build/share “asset health”


dashboards with plant management
• Extra integration effort to do in PI
• Extra software licensing and training to
do in native CM environment

#PIWorld ©2019 OSIsoft, LLC 12


Solution: PUT THE DATA INTO PI!
SETPOINT® Edge Device and
Machines PI System
Machinery Protection System

ALL*
Vibration
Vibration
Sensors
Data

*Includes high-speed
waveform data

#PIWorld ©2019 OSIsoft, LLC 13


Vibration Monitoring Systems (As-Left)
Unit Machine Size Sensors MPS CMS

1 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT


2 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT
3 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT
4 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT
5 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT
6 GE Frame 7B* 60MW Proximity SETPOINT SETPOINT
7 GE steam turbine 155MW Proximity BN3500 SETPOINT
8 Alstom GT10B2* 25MW Proximity BN3500 None
Integrates machinery
* Combustion turbines protection and condition
monitoring functions

#PIWorld ©2019 OSIsoft, LLC 14


Enables

1. Online ConMon = never miss important data


• Specialist analyzes data, rather than collects data
• Proactive – doesn’t matter if not there when problems occur
• i-factor technology detects change automatically and does
not rely on human-intensive alarm setting/management

ALL DATA PRESENT


AND ACCOUNTED FOR!
#PIWorld ©2019 OSIsoft, LLC 15
Enables

2. Remote access
• Data is accessible remotely and securely to offsite specialist
• Can access in minutes instead of hours
• Move data, not people

#PIWorld ©2019 OSIsoft, LLC 16


Enables
3. Integrated data
• Uses PI as repository for all
data
• Process and vibration is in one PROCESS
place; correlation and DATA
cause/effect more easily VIBRATION
ascertained +
DATA
• Reduced IT burden VIBRATION
(1 system) DATA
• In PI = Accessible to other
analytic tools such as APR

#PIWorld ©2019 OSIsoft, LLC 17


Enables

4. Viewer software is PI and special


free download
• Open (.cms) file format similar in concept to
PDF
• Easy to share data when/where needed
• No extra licenses or fees
• Enhances collaboration to solve problems

#PIWorld ©2019 OSIsoft, LLC 18


Enables

5. Easily build/share “asset health”


dashboards with plant management
• No special integration effort to do in PI
• Can drill down to special visualization tool when required
to do deep diagnostic analysis

#PIWorld ©2019 OSIsoft, LLC 19


DEMO
Sample Screens for Anomaly Detection

#PIWorld ©2019 OSIsoft, LLC 20


Compare cold starts
using event frames

PI Vision screen w/
bearing amp / phase data

#PIWorld ©2019 OSIsoft, LLC 21


PI Vision screen overlaying
cold start events

Drill down even further into CMS Display


app to view WAVEFORMS and specialized plots
in region of interest (e.g. startup)

#PIWorld ©2019 OSIsoft, LLC 22


Now in CMS Display app
w/ “deep dive” tools

Full Spectrum Cascade


Orbit / Timebase (frequency domain waveforms)
(time domain waveforms)

Time slider
with multi-variable trends
(80ms resolution)
#PIWorld ©2019 OSIsoft, LLC 23
Now in CMS Display app
w/ “deep dive” tools

Bode Plot
(amp/phase vs. speed)

Polar Plot
Detailed profile showing startup, (amp/phase vs. speed)
steady-state, and shutdown

#PIWorld ©2019 OSIsoft, LLC 24


Portland General Electric
Putting comprehensive vibration data into the PI System

CHALLENGE SOLUTION RESULTS

No online condition monitoring SETPOINT® Edge Device Overall company risk


data when machinery decisions streamed all vibration data into significantly reduced.
needed to be made. existing PI System
• Eliminated separate infrastructure;
• Incumbent system required used PI System already installed • Can now identify critical asset
expensive additional infrastructure • All data in same “system of record” for health accurately
ease of correlation • We know when we’re healthy (and
• Offline approach was too reactive
and involved 10-hour roundtrip • Accessible to everyone in the plant when we’re sick)
travel to/from site • WAVEFORMS IN PI! • Data analysis done in 1 hour
(the key to accurate analysis) instead of 15+ hours

#PIWorld ©2019 OSIsoft, LLC 25


Questions? Please remember

Please wait for


the microphone
DOWNLOAD
THE MOBILE
State your
APP
name & company

#PIWorld ©2019 OSIsoft, LLC 26


Your presenters
• Austin Curtis
• Maintenance Manager
• Portland General Electric
[email protected]
• Steve Sabin
• VP – Marketing & Product Management
• Brüel & Kjær Vibro
[email protected]
#PIWorld ©2019 OSIsoft, LLC 27
#PIWorld ©2019 OSIsoft, LLC 28

You might also like