Post on 30-Aug-2020
zhaw
Deep Learning for Fault Detection and Isolation
Hierarchical Networks for PHM
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 1
Gabriel Michau1, Thomas Palmé2, Olga Fink1
1 - Zurich University of Applied Science, Switzerland2 - General Electric (GE) Switzerland
Smart Maintenance Conference 2018
Zürich
zhaw
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 2
Prognostics and Health Management
Data Acquisition
Data Processing
Condition assessment(detection)
Diagnostics(identification)
Prognostics
Decision
• Features
• Raise early alarm
• Provide clear indication on the cause
Automatic process in a single approach
zhaw
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 3
Prognostics and Health Management
Data Acquisition
Data Processing
Condition assessment(detection)
Diagnostics(identification)
Prognostics
Decision
zhawChallenges of Critial Systems
PHM for critical (and complex) systems• Faults:
• Faults are rare • Faults cannot be afforded → preventive maintenance• Possible faults are numerous (possibly hitherto unknown)• Consequences of faults can be diverse
• System:• Heterogeneous data• Varied operating condition over long time scale• Unit Specificity
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 4
zhawHierarchical Neural Networks
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 5
“Health Indicator”
Trained with healthy condition monitoring data only
zhawTest in real case studies
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 6
Healthy
Abnormal behavior 100 days before!
Generator with inter-turn failure
zhaw
Water Temperature Shaft Voltage Rotor Flux Rotor Flux
In agreement with • other models (AAKR, PCA) • with expert knowledge.
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 7
At no additional cost!
Generator Diagnostics
zhawNew Challenges
What if, I don’t have enough training data?• What if my system is new (or has been refurbished)?• What if I am expecting operating conditions to change?
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 8
⇒Use data from other similar systems
Fleet approach to PHM (from manufacturer perspective)
zhawExample with a stator (vane failure)
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 9
Long training Short training
Winter
zhawHealthy Stator
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 10
Summer
Long training Short training
zhawSubfleet selection
• What data do I need?• Not everything at once because
1. It is too much2. It will contains operating points that are not relevant to my unit3. Completely different operating conditions might hide faults
⇒I need data relevant for my unit.
• Within a (manufacturer) fleet, all units are similar. ⇒Need to identify other units with similar operating conditions
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 11
zhawSubfleet selection
• Multi-dimensional datasets comparison• Resource and time consuming (multi-dimensional distance and distributions)• Difficult (number of nearest neighbors evolves as n.D)
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 12
zhawSubfleet selection
• Solution: The Hierarchical Network has proven to be efficient in measuring how well the test data correspond to the training data.
⇒Idea: 1. Train HELM on one unit, test with data from other units2. select those that are detected as similar.
⇒Computationally efficient: • HELM takes all dimensions and output a single indicator• 1 HELM to train per unit (instead of N2 dataset comparisons)
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 13
zhawProposed Methodology
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 14
HELMNew Stator “Healthy”
Training
1 – Train HELM on “new” stator
zhawProposed Methodology
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 15
TrainedHELM
New Stator
Test
Healthy Stator
1 – Train HELM on “new” stator2 – Test all other Stator
zhawProposed Methodology
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 16
HELMHealthy Stator “Healthy”
Training
1 – Train HELM on “new” stator2 – Test all other Stator
3 – Train HELM with other stator
zhawProposed Methodology
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 17
TrainedHELM
Healthy Stator
Test
New Stator
1 – Train HELM on “new” stator2 – Test all other Stator
3 – Train HELM with other stator4 – Test with “new” stator
zhawProposed Methodology
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 18
1 – Train HELM on “new” stator2 – Test all other Stator
3 – Train HELM with other stator4 – Test with “new” stator
5 – Dissimilarity measure:
Healthy StatorNew Stator
HELM
zhawSimilar datasets
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 19
Similar dataset if d below a threshold
Need to allow for more than 20% of dissimilar point to find onesimilar dataset for half of the dataset
This fleet has very dissimilar unitsDifficulty to find subfleets
zhawIndividual Asset Monitoring
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 20
HELMHealthy Stator
Subfleet “Healthy”
Training
New Stator
1 – Train HELM on “new” stator subfleet
zhawIndividual Asset Monitoring
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 21
trainedHELM
1 – Train HELM on “new” stator subfleet
2 – Monitor new data from individual asset
Test
New StatorNew data Health
Healthy StatorSubfleet
New Stator
zhawExamples
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 22
zhawExamples
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 23
zhawExamples
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 24
zhawExamples
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 25
zhawWhat is coming next?
• Fleet Feature analysis: What can be learned on the fleet?
• Transfer learning: What if my fleet have several versions of my system? (eg., not exactly the same data recorded)
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 26
zhaw
Thanks for your attention
04.09.2018 SMC 2018 – © ZHAW – Gabriel Michau 27
Acknowledgements