Using search for engineering diagnostics and prognostics

1
Using search for engineering diagnostics and
prognostics

• Jim Austin

2
Overview

• Problem domain
• Drivers - why we need better solutions
• Example applications
• Our approach
• Challenges

3
Prognostics and Diagnostics

• Find out what is wrong with some thing
• Find out what may be about to happen
• Use data to achieve this, but deliver knowledge
• Wide applicability (not just engineering)

4
Engineering problems

• Asset monitoring
• Large numbers of sensors
• Many types of sensors
• Distributed sensors and systems
• Possibly hostile domains
• Large data rates
• Slow connections
• Data incomplete, noisy hard to characterise

5
Engineering problems

• Response
• Needs to be rapid
• Qualified response (i.e. how good)
• Must include users in the loop, not yet automatic
• Conclusion must be justified able to dig into
  problem

6
Drivers

• Why now?
• Sensors are now robust small and reliable
• Data collection is very cost effective (2Tb lt
  200)
• Large computing capability is now possible
• Data to Knowledge is a prime motivator
• Most easy wins have been achieved
• Green agenda is forcing issues

7
Example Applications

• Aero-engines fixed assets
• Rail track and carrage
• Roads

8
Gas turbines

• High speed, rotating systems
• Typically very reliable
• Used for air travel as well as pumps and
  generators (oil and gas, marine, air, power)

9
Gas turbines

• Typical problem
• Spot failure in good time (!)
• Spot maintenance issue ahead of time
• Data is
• High frequency
• Large
• Complex

10
Rail

• Monitoring of both track and carriages
• Over 2000 alerts on a Thomas virgin voyager
• Aim is to reduce unplanned maintenance

11
Rail

• Track
• Look at data from track inspection systems
• Find if track is bent or broken and needs
  maintenance

12
Road

• Monitoring for congestion problems
• Data from road loops (flow and occupancy)
• Weather
• Accident reports
• Adjust
• Traffic lights
• Variable message signs

13
Road
Hull road bus gate, York
14
Road
15
Our approach

• Use historic data as a prediction of now and the
  future
• Basically search the historic data
• Use AURA neural network
• Have a set of systems within Signal Data Explorer
• Share data and services through portals
• Building on CARMEN

16
SDE and CARMEN
17
Data compatibility

• Neural Data Format NDF
• Allow interoperability between
• Multi-channel systems data (.mcd).
• Comma delimited (.csv).
• Alpha map (.map).
• Neural event (.nev).
• NeuroShare native (.nsn).
• Nex (.nex).
• PC spike2 (.smr).
• Plexon data (.plx).
• TDT data format (.stb)
• Supported in visualisation tool (SDE), soon in
  services

18
Data entry
19
Services
20
Execution log
21
Examples
22
Search for signals
Correlation matrix
Data converter
Time series data
Historical data
Compare
Known?
23
Fault identification
Correlation matrix
Data converter
Time series data
Historical data
Compare
Known?
24
Challenges

• Best practice in data collection build system
  when you know how to process it!
• Better tools, for analysis of signals, images and
  text (three main groups).
• Better collaborative technologies, new in
  industry sector
• User adoption of the technology

25
Summary

• Data now available in large quantities
• Real opportunities to improve the systems that
  are being built