The document discusses gradient-based fault isolation techniques. It begins by explaining the need for both fault detection and fault isolation in industrial processes. It then discusses commonly used residual-based approaches for fault detection and introduces the motivation for gradient-based fault isolation approaches. These approaches use partial derivatives of models to identify the process variables most responsible for detected faults. The document outlines some tools for visualizing fault isolation results and discusses opportunities to improve gradient aggregation methods and further narrow isolation performance gaps. It concludes that gradient-based techniques can perform fault isolation without prior fault information and more development is still needed to strengthen results.

1. Gradient-based Fault Isolation
Residual-based Fault Detection Systems
Francisco Serdio Fernández
Department of Knowledge-Based Mathematical Systems
Johannes Kepler University Linz, Austria
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
for
http://www.flll.Francisco Serdio jku.at/staff/francisco

2.  Why Fault Detection (FD) ?
 Why Fault Isolation (FI) ?
 FD with Residual-based approaches
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

3. WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Why Fault Detection?
http://www.flll.Francisco Serdio jku.at/staff/francisco

4. Why Fault Detection?
 Products with high quality demands
 High quality is required also in the production chain
 High quality is required also in the supply chain
[1] D. Blanchard. Supply Chain Management Best Practices. John Wiley & Sons,
Hoboken, NJ, USA, 2007.
 Continuity in the production lines
 Minimum down-time
[2] R. Iserman. Fault-Diagnosis Applications. Model-Based Condition Monitoring:
Actuators, Drives, Machinery, Plants, Sensors, and Fault-tolerant Systems. Springer,
Berlin Heidelberg, Germany, 2011.
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 High quality processes imply
http://www.flll.Francisco Serdio jku.at/staff/francisco

5. WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Why Fault Detection?
http://www.flll.Francisco Serdio jku.at/staff/francisco

6.  Manual supervision is not affordable or in some
cases simply impossible
 The precision of manual supervision usually depends
 on the experience of the operators
 and even on their performance on a given day
[3] E. Lughofer, J.E. Smith, P. Caleb-Solly, M. Tahir, C. Eitzinger, D. Sannen and M.
Nuttin. (2009). Human-machine interaction issues in quality control based on on-line
image classication. IEEE Transactions on Systems, Man and Cybernetics, part A:
Systems and Humans, 39(5), 960-971.
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Why Fault Detection?
 Manual process supervision
http://www.flll.Francisco Serdio jku.at/staff/francisco

7.  Why Fault Detection (FD) ?
 Why Fault Isolation (FI) ?
 FD with Residual-based approaches
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

8. Fault Isolation
Detection
Needle
Needle !!
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Why Fault Isolation?
Haystack
http://www.flll.Francisco Serdio jku.at/staff/francisco

9. Why Fault Isolation?
 Multiple sensor networks turned out to emerge
in industrial settings and factories
 Huge amount of sensors and actuators to check
 Manual supervision is not affordable or in some
cases simply impossible
 Any valuable information regarding where the fault
is
located could be a great aid for the operator
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
  Isolation !
http://www.flll.Francisco Serdio jku.at/staff/francisco

10.  Why Fault Detection (FD) ?
 Why Fault Isolation (FI) ?
 FD with Residual-based approaches
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

11. FD with Residual-based approaches
 Algebraic relationships among different sensors
 Difference relationships among different sensor
outputs and actuator inputs
 Inconsistencies, expressed as residuals, can be
used for detection and isolation purposes
[4] V. Venkatasubramanian, R. Rengaswamy, S. Kavuri and K. Yin. (2003). A review of
process fault detection and diagnosis: Part iii: Process history based methods.
Computers & Chemical Engineering, 27(3), 327-346.
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 Analytical Redundancy
 Direct redundancy
 Temporal redundancy
http://www.flll.Francisco Serdio jku.at/staff/francisco

12. FD with Residual-based approaches
Analytical Redundancy graphically
Moving from the signal space to the residual space: illustrating an untypical signal pattern
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

13. Tracking residuals within a dynamic tolerance band
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

14. Recall FD with Residual-based
approaches
 More information regarding Fault Detection in
[5] F. Serdio, E. Lughofer, K. Pichler, T. Buchegger and H. Efendic, Data-Driven Residual-Based
Fault Detection for Condition Monitoring in Rolling Mills. Proceedings of the IFAC Conference on
Manufacturing Modeling, Management and Control, MIM '2013, St. Petersburg, Russia, 2013, pp.
1546-1551. (Winner of MIM 2013 Best paper award)
[6] F. Serdio, E. Lughofer, K. Pichler, T. Buchegger, and H. Efendic, Residual-based Fault Detection
using Soft Computing techniques for Condition Monitoring at Rolling Mills. Information Sciences,
vol. 259, pp. 304–330, 2014.
[7] F. Serdio, E. Lughofer, K. Pichler, T. Buchegger, M. Pichler and H. Efendic, Multivariate Fault
Detection using Vector Autoregressive Moving Average and Orthogonal Transformation in the
residual Space. Annual Conference of the Prognostics and Health Management Society, PHM 2013,
New Orleans, LA, USA, 2013, pp. 548-555.
[8] F. Serdio, E. Lughofer, K. Pichler, T. Buchegger, and H. Efendic, Fault Detection in Multisensor
Networks based on Multivariate Time-series Models and Orthogonal Transformations.
Information Fusion, vol. under revision (minor), 2014.
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

15.  Why Fault Detection (FD) ?
 Recall FD with Residual-based approaches
 Why Fault Isolation (FI) ?
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

16. Motivation of the FI Gradient-based
approaches
 We are blind about faults
 We do not know how a fault looks like
 We do not have fault patterns (labeled data)
 Process variable contribution plot
 There is an extension to non-linear PCA
 It reverts back to the original process variables
[9] P. Miller, R. Swanson, and C. Heckler, Contribution plots: A missing link in multivariate quality
control. Applied Mathematics and Computer Science, vol. 8, p. 775792, 1998.
[10] F. Jia, E. Martin, and A. Morris, Nonlinear principal components analysis with application to
process fault detection. International Journal of Systems Science, vol. 31, p. 14731487, 2001.
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 There is literature about PCA
http://www.flll.Francisco Serdio jku.at/staff/francisco

17. Motivation of the FI Gradient-based
approaches
 Partial derivatives !
 With respect to a specific dimension can indicate the
relative importance of the corresponding variable
(channel) on that function
 Can be computed according to the model expression
 Can be computed by means of numeric tricks
 We can plug a FI system to any FD model !
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

18. Motivation of the FI Gradient-based
approaches
 How do we revert back to the original process
variables?
 We compute the gradients of the model variables
 We aggregate the gradients
 We get a candidate responsible variable
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 We take the warning models
 Crisp decision
 Fuzzy decision
http://www.flll.Francisco Serdio jku.at/staff/francisco

19. Aggregating gradients
 Biggest gradient as faulty channel
 A channel is either (properly) isolated or not
 Several channels are proposed as faulty
 There are normalized against the channel with the
highest gradient aggregation
 By definition, it will produce always better results than
its crisp counterpart
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 Crisp decision
 Winner takes all approach
 Fuzzy decision
http://www.flll.Francisco Serdio jku.at/staff/francisco

20.  Why Fault Detection (FD) ?
 FD with Residual-based approaches
 Why Fault Isolation (FI) ?
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

21. Tools to depict Fault Isolation
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

22.  Why Fault Detection (FD) ?
 FD with Residual-based approaches
 Why Fault Isolation (FI) ?
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

23. WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Results
http://www.flll.Francisco Serdio jku.at/staff/francisco

24. WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
Results
http://www.flll.Francisco Serdio jku.at/staff/francisco

25.  Why Fault Detection (FD) ?
 FD with Residual-based approaches
 Why Fault Isolation (FI) ?
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

26. Can do we more ?
 We must work in how to aggregate the
gradients
 Weight the gradients with other data
 We are using violation degree of the threshold
 We are using quality of the model
 Goal: narrow the Fault Isolation Gap (FIG)
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 Time frames (sliding windows)
http://www.flll.Francisco Serdio jku.at/staff/francisco

27.  Why Fault Detection (FD) ?
 FD with Residual-based approaches
 Why Fault Isolation (FI) ?
 Motivation of the FI Gradient-based approaches
 Tools to depict Fault Isolation
 Results
 Can do we more ?
 Conclusions
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
http://www.flll.Francisco Serdio jku.at/staff/francisco

28. Conclusions
 We can perform Fault Isolation (FI) without
information about the faults
 Only based on warning models and gradients
 We have introduced new tools to depict FI
 We must still strength the results
WCCI 2014 / July 6-11 / Beijing, China
francisco.serdio@jku.at
 Graphically
 Numerically
http://www.flll.Francisco Serdio jku.at/staff/francisco

29. Thanks a lot for your attention!
WCCI 2014 / July 6-11 / Beijing, China
{francisco.serdio,edwin.lughofer}@jku.at
http://www.flll.jku.at/staff/{Francisco Serdio, Dr. Edwin Lughofer francisco,lughofer}