Deep Learning for Fault Detection and Isolation...Prognostics and Health Management Data Acquisition...

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

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

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

zhawHierarchical Neural Networks

“Health Indicator”

Trained with healthy condition monitoring data only

zhawTest in real case studies

Healthy

Abnormal behavior 100 days before!

Generator with inter-turn failure

Water Temperature Shaft Voltage Rotor Flux Rotor Flux

In agreement with • other models (AAKR, PCA) • with expert knowledge.

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?

⇒Use data from other similar systems

Fleet approach to PHM (from manufacturer perspective)

zhawExample with a stator (vane failure)

Long training Short training

Winter

zhawHealthy Stator

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

• Multi-dimensional datasets comparison• Resource and time consuming (multi-dimensional distance and distributions)• Difficult (number of nearest neighbors evolves as n.D)

• 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)

zhawProposed Methodology

HELMNew Stator “Healthy”

Training

1 – Train HELM on “new” stator

TrainedHELM

New Stator

Healthy Stator

1 – Train HELM on “new” stator2 – Test all other Stator

HELMHealthy Stator “Healthy”

Training

3 – Train HELM with other stator

TrainedHELM

Healthy Stator

New Stator

3 – Train HELM with other stator4 – Test with “new” stator

5 – Dissimilarity measure:

Healthy StatorNew Stator

zhawSimilar datasets

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

HELMHealthy Stator

Subfleet “Healthy”

Training

New Stator

1 – Train HELM on “new” stator subfleet

zhawIndividual Asset Monitoring

trainedHELM

1 – Train HELM on “new” stator subfleet

2 – Monitor new data from individual asset

New StatorNew data Health

Healthy StatorSubfleet

New Stator

zhawExamples

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)

Thanks for your attention

Acknowledgements

Deep Learning for Fault Detection and Isolation...Prognostics and Health Management Data Acquisition...

Documents

Transcript of Deep Learning for Fault Detection and Isolation...Prognostics and Health Management Data Acquisition...

Probabilistic Fatigue Damage Diagnostics and Prognostics...probabilistic delamination size and location detection framework using Bayesian imaging method (BIM) is proposed and validated

Prognostics and Health Management - F-16

Prognostics & Vehicle Health Management in the Automotive ... · Predictive Analytics. Big Data & Cloud Computing. Telematics & Fast Networks 4G + ... Combining Physics-Based & Data-Driven

Prognostics & Health Management at GE & Health Management at GE Dr. Piero P.Bonissone Industrial AI Lab GE Global Research NCD Select detection model Anomaly detection results presentation

Machinery Prognostics and Health Management · component and machine level prognostics and diagnostic research, there is very little prognostics or diagnostic research work that considers

Bearing Prognostics Simulator (BPS)

Guide for Hull Condition Monitoring Systems · operational data, sensor data and other supporting data to provide structural health diagnostics and , prognostics through the correlation

Options for Prognostics Methods: A review of data - PHM Society

Integrated Diagnostics and Prognostics Rotating Machinery

A Data-Driven Failure Prognostics Method.pdf

Local-Level Prognostics Health Management Systems ...

Prognostics Assessment Using Fleet-wide Ontology

Chapter 1. Introduction : Prognostics & Health Management

Actuator Systems Prognostics - NASA · Actuator Systems Prognostics Edward Balaban, Abhinav Saxena, Prasun Bansal, Kai Goebel Prognostics Center of Excellence, NASA ARC Overview Challenges

Prognostics in Data Centers - web.stanford.edu · Runaway Process Detection • R-cloud construction using a linear separating primitive in two dimensions Primitive Bundle R-cloud

Fault Detection and Classification of Time Series …Time series data generated by sensors in physical systems and data mining on such data are widely used in applications of Prognostics

Prognostics and Health Management for …...International Journal of Prognostics and Health Management, ISSN 2153-2648, 2017 060 1 Prognostics and Health Management for Maintenance

Filter Clogging Data Collection for Prognostics - PHM … · Filter Clogging Data Collection for Prognostics O. F. Eker1, F. Camci1, and I. K. Jennions1 1 IVHM Centre, Cranfield University,

Prognostics Overview Electronic Prognostics · defense, military, medical, and other sectors. To ensure high levels of reliability and performance, prognostics technology developers

Options for Prognostics Methods: A review of data-driven and physics … · 2013-10-09 · Options for Prognostics Methods: A review of data-driven and physics- ... The goal of this