1. Proceedings
  2. I3M
  3. 2019
  4. IMAACA
  5. 10.46354/i3m.2019.imaaca.001

A combined bond graph-based – data-based approach to failure prognosis

  • a Wolfgang Borutzkyi  
  • aBonn-Rhein-Sieg University of Applied Sciences, St. Augustin, Germany
Cite as
Borutzky W. (2019). A combined bond graph-based – data-based approach to failure prognosis. Proceedings of the 12th International Conference on Integrated Modeling and Analysis in Applied Control and Automation (IMAACA 2019), pp. 1-10. DOI: https://doi.org/10.46354/i3m.2019.imaaca.001

Abstract

Given known control inputs and real sensor outputs or simulated measurements, the paper shows that numerical values of unknown parameter degradation functions can be obtained by evaluating equations derived from a bicausal diagnostic bond graph that are not analytical redundancy relations. Inspection of causal paths beforehand enables to decide whether potential parametric faults can be isolated with a number of sensors in given locations. The proposed approach can be applied in the case of multiple isolated simultaneous parametric faults. Numerical values of degradation functions can be computed concurrently to the constant monitoring of a system and the measurement of signals. Repeatedly projecting the time evolution of a degradation function into the future based on values in a sliding time window enable to obtain a sequence of remaining useful life estimates. The novel proposed combined bond graph-model-based, data-based approach is verified by an offline simulation study of a typical power electronic circuit."

References

  1. Alem S, Benazzouz D (2013) Optimal sensor placement for detection and isolation by the structural adjacency matrix. Int J of the Physical Sciences 8(6):225–230, DOI 10.5897/IJPS12.547
  2. Benmoussa S, Ould Bouamama B, Merzouki R (2014) Bond graph approach0 for plant fault detection and isolation: Application to intelligent autonomous vehicle. IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING 11(2):585–593
  3. BorutzkyW(ed) (2016) Bond Graphs for Modelling, Control and Fault Diagnosis of Engineering Systems, 2nd edn. Springer International Publishing Switzerland, DOI 10.1007/978-3-319-47434-2
  4. BorutzkyW(2018a) Determination of a function for a degradation process by means of two diagnostic bond graphs. In: Proceedings of the 10th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, IFAC, Warsaw, Poland
  5. BorutzkyW(2018b) Sensor Placement on Diagnostic Bond Graphs For Maximum Structural Isolation of Parametric Faults. In: Granda J, Karnopp D (eds) Proceedings of the 13th International Conference on Bond Graph Modeling and Simulation (ICBGM’2018), SCS, Bordeaux, France, Simulation Series, vol 50(12), pp 41–49
  6. Chi G, Wang D (2015) Sensor Placement for Fault Isolability Based on Bond Graphs. IEEE Trans on Automatic Control 60(11):3041–3046
  7. Djeziri M, Ould Bouamama B, Merzouki R, Dauphin-Tanguy G (2009) Optimal sensor placement for fault diagnosis. In: 2009 IEEE International Conference on Mechatronics, Malaga, Spain, 2009
  8. Escobar LA, Meeker WQ (2006) A review of accelerated test models. Statistical Science 21(4):552 – 577, DOI 10.1214/ 088342306000000321
  9. Frisk E, Krysander M, Åslund J (2009) Sensor placement for fault isolation in linear differential-algebraic systems. Automatica 45:364 – 371
  10. Gawthrop P (1995) Bicausal Bond Graphs. In: Cellier F, Granda J (eds) ICBGM’95, International Conference on Bond Graph Modeling and Simulation, SCS Publishing, Simulation Series, vol 27(1), pp 83–88
  11. Jha M (2015) Diagnostics and prognostics of uncertain dynamical systems in a bond graph framework. PhD Thesis, École Centrale de Lille, Université Lille Nord-de-France
  12. Kulkarni C, Biswas G, Koutsoukos X, Goebel K, Celaya J (2010) Physics of failure models for capacitor degradation in DC-DC converters. In: The Maintenance and Reliability Conference, MARCON, Knoxville, TN, U.S.A.
  13. Medjaher, K, Zerhouni, N (2013) Hybrid Prognostic Method Applied to Mechatronic Systems. International Journal of Advanced Manufacturing Technology 69(1–4):823–834, DOI 10.1007/
    s00170-013-5064-0
  14. Prakash O, Samantaray A, Bhattacharyya R, Ghoshal S (2018) Adaptive prognosis for a multi-component dynamical system of unknown degradation modes. In: Proceedings of the 10th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes„
    IFAC, Warsaw, Poland
  15. Samantaray A, Ghoshal S (2008) Bicausal bond graphs for supervision: From fault detection and isolation to fault accommodation. Journal of the Franklin Institute 345:1–28
  16. Sankararaman S, Goebel K (2013) Why is the Remaining Useful Life Prediction Uncertain ? In: Annual Conference of the Prognostics and Health Management Society, pp 1 – 13, open-access article
  17. Savitzky, A, Golay, MJE (1964) Smoothing and differentiation of data by simplified least squares procedures. Analytical Chemistry 38(8):1627 – 1639
  18. The Mathworks (n.d.) Three ways to estimate remaining useful life. White paper, https://www.mathworks.com/products/ predictive-maintenance.html