Friday , April 26 2024

Monitoring of a Milk Manufacturing Workshop Using Chronicle and Fault Tree Approaches

Anis M’HALLA
LAboratoire de Recherche en Automatique, Ecole Nationale d’Ingénieurs de Tunis
BP 37, le Belvédère, 1002 Tunis, TUNISIE

Simon Collart DUTILLEUL 
Laboratoire d’Automatique, Génie Informatique et Signal, Ecole Centrale de Lille, Cité Scientifique
BP 48, 59651 Villeneuve d’Ascq, FRANCE

Etienne CRAYE
Laboratoire d’Automatique, Génie Informatique et Signal Ecole Centrale de Lille, Cité Scientifique
BP 48, 59651 Villeneuve d’Ascq, FRANCE

Mohamed BENREJEB
LAboratoire de Recherche en Automatique, Ecole Nationale d’Ingénieurs de Tunis
BP 37, le Belvédère, 1002 Tunis, Tunisie

Abstract: Developments presented in this paper are devoted to the monitoring of manufacturing job-shops with time constraints and with assembling tasks. A new method for monitoring combining chronicles with fault tree analysis is proposed. The purpose of the proposed approach is to explain in details what is happening on the system and to help operators identifying failures in order to avoid a damage of the process or an accident with human beings. Finally, to demonstrate the effectiveness and accuracy of the monitoring approach, an application to a milk production unit is outlined. The results show that the monitoring approach allows keeping on producing, by on-line diagnosis, while providing correct quality of the manufactured products.

Keywords: Milk manufacturing unit, monitoring, chronicle, fault tree, diagnosis, time constraints

>>Full text
CITE THIS PAPER AS:
Anis M’HALLA, Simon Collart DUTILLEUL, Etienne CRAYE, Mohamed BENREJEB, Monitoring of a Milk Manufacturing Workshop Using Chronicle and Fault Tree Approaches, Studies in Informatics and Control, ISSN 1220-1766, vol. 19 (4), pp. 379-390, 2010.

1. Introduction

Supervision and monitoring are a set of solutions allowing managing the process in order to correctly react in case of failure. The problem we deal with is the supervision of complex discrete event systems such as telecommunication networks, electricity distribution networks and manufacturing workshops. The Manufacturing systems can also be subject to staying time constraints (maximum cooking time of a product in a furnace, overheating of milk bottles in a hydromat …). This type of constraints does not only affect the performances of the system but also its functional validity (burnt pieces, unusable milk…). Hence, we need a specific performance evaluation, because an “as-soon-as-possible” operation is invalid in the general case.

Two classical approaches in monitoring such systems are knowledge based techniques that directly associate a diagnosis to a set of symptoms, for example expert systems [1], or chronicle recognition systems [2], and model-based techniques which rely on a behavioural model of the system [3]. The main weakness of the second approach is the difficulty to represent the behavioral model of complex systems. Therefore, we focus on expertise-based approaches which are known to be better suited to that kind of system than model-based techniques.

In this paper, we propose a new method for monitoring manufacturing workshops with time constraints which combine chronicles with fault tree analysis. The monitoring system is used as a model for diagnosis. The proposed approach allows to roughly identify what is happening on the system (failed and critical components) and to help operators to identify failures in order to avoid damage to the equipment or an accident with human beings.

A fault tree describes how a set of events can concur in order to cause a top event. However, to address the problem of detection, we complement those fault trees with a dynamic model of system behaviour. This system can capture the behavioural transformations that occur in complex systems as a hierarchy of state machines, by chronicle recognition. The chronicle, proposed in [4], describes a situation that is worth identifying within the diagnosis context. It is made up of a set of events and temporal constraints between those events. As a consequence, this formalism fits particularly well problems that have a temporal dimension. Then, monitoring the system consists of analyzing flows of events and recognizing constraints composing the chronicles.

The work has been motivated by an application that aims at monitoring the behaviour of a milk manufacturing workshop. In such system, each operation is associated to a time interval. Its lower bound indicates the minimum time needed to execute the operation. The upper bound sets the maximum time to not exceed otherwise the quality of the product is deteriorated.

This paper is organised as follows. In Section 2 we discuss the form of the monitoring model and its development process and we recall the principles of the chronicle recognition approach. A more detailed description of this distributed approach is given in [4], [5] and [14]. Section 3 begins by presenting the milk production workshop. The monitoring approach of manufacturing systems is considered. An illustrative example is outlined and the results are discussed. Finally, conclusions of this work are given.

References:

  1. Niebur, D., Expert Systems for Power System Control in Western Europe. IEEE Symp. on Intel. Control, Philadelphia, 1990.
  2. Cordier, M. O., C. Dousson, Alarm Driven Monitoring Based on Chronicles. 4th Symp. on Fault Detection, Supervision and Safety for Technical (Safeprocess), Budapest, 2000.
  3. Riese, M., Diagnosis of Extended Finite Automata as a Dynamic Constraint Satisfaction Problem. Intl. Workshop on Principles of Diagnosis (DX-93), Aberystwyth, 1993.
  4. Dousson, C., P. Gaborit, M. Ghallab, Situation Recognition: Representation and Algorithms, Intl. Joint Conf. on Artificial Intel. (IJCAI’93), Chambéry, 1993, pp. 166-172.
  5. Cordier, M. O., X. Le Guillou, S. Robin, L. Rozé, T. Vidal, Distributed Chronicles for On-line Diagnosis of Web Services, 18th Intl. Workshop On Principles of Diagnosis (DX’07), Nashville, May 2007, pp. 37-44.
  6. Pencolé, Y., M. O. Cordier, A Formal Framework for the Decentralised Diagnosis of Large Scale Discrete Event Systems and Its Application to Telecommunication Networks. Artificial Intel. Journal, Vol. 164(1-2), 2005, pp. 121-170.
  7. M’halla, A., N. Jerbi, S. Collart Dutilleul, E. Craye, M. Benrejeb, Fuzzy Filtering of Sensors Signals in Manufacturing Systems with Time Constraints. Intl. Journal of Computers, Comm. & Control (IJCCC). Vol. 5(3), 2010, pp. 362-374.
  8. Silveira, D., M. Combacau, A. Boufaied, Prognosis and Recovery Evaluation in Flexible Manufacturing Supervision. Journal of Decision Systems, Vol. 12, 2003, pp. 93-109.
  9. Guillou, X., M. O. Cordier, S. Robin, l. Roze, Chronicles for On-line Diagnosis of Distributed Systems. 18th European Conf. on Artificial Intel. (ECAI’08), Patras, July 2008.
  10. Dousson, C., P. Le Maigat, Chronicle Recognition Improvement Using Temporal Focusing and Hierarchization. Proc. of the 20th Intl. Joint Conf. on Artificial Intel. (IJCAI), Hyderabad, January 2007.
  11. Taisne, J., Intelligent Alarm Process for DMS Based on Chronicle Concept. 19th Intl. Conf. on Electricity Distribution (C I R E D), Vienna, May 2007.
  12. Boufaied, A., A. Subias, M. Combacau, Distributed Time Constraints Verification Modelled with Time Petri Nets. 17th IMACS World Cong. Scientific Computation, Applied Math. and Simulation, Paris, July 2005.
  13. Boufaied, A., A. Subias, M. Combacau, Détection distribuée par reconnaissance floue de chroniques. Journal Européen des Syst. Aut. (JESA), Vol. 40(2), 2006, pp. 233-259.
  14. Mhalla, A., N. Jerbi, S. Collart Dutilleul, E. Craye, M. Benrejeb, Distributed Monitoring Based on Chronicles Recognition for Milk Manufacturing Unit. Journal. of Aut. & Syst. Eng. (JASE), Vol. 4(1), 2010.
  15. Andrews, J. D., T. R. Moss, Reliability and Risk Assessment, 2nd ed., Prof. Eng. Publishing Limited, Bury St Edmunds and London, 2002.
  16. Enrico, Z., F. D. Francesco, Processing Dynamic Scenarios from a Reliability Analysis of a Nuclear Power Plant Digital Instrumentation and Control System. Journal of Annals of Nuclear Energy, Vol. 36(9), 2009, pp. 1386-1399.
  17. Yuhua, D., Y. Datao, Estimation of Failure Probability of Oil and Gas Transmission Pipelines by Fuzzy Fault Tree Analysis. Journal of Loss Prevention in the Process Industries, Vol. 18(2), 2005, pp. 83-88.
  18. Chen, J., J. Howell, A Self-validating Control System Based Approach to Plant Fault Detection and Diagnosis. Computers & Chemical Engineering, Vol. 25(2-3), 2001, pp. 337-358.
  19. Ericson, C. A., Fault Tree Analysis – a History. 17th Intl. System Safety Conf., Orlando, USA, 1999.
  20. Kaiser, B., P. Liggesmeyer, O. Mackel, A New Component Concept for Fault Trees. 8th Australian Workshop on Safety Critical Systems and Software, Canberra, 2003.
  21. Cordier, M. O, X. Le Guillou, S. Robin, L. Rozé, T. Vidal, Distributed Chronicles for On-line Diagnosis of Web Services. 18th Intl. Workshop On Principles of Diagnosis (DX’07), Nashville, May 2007 pp. 37-44.
  22. Morin, B., D. Herve, Correlation of Intrusion Symptoms: An Application of Chronicles. 6th Intl. Sym. on Recent Advances in Intrusion Detection, Pittsburgh, 2003.
  23. Sallak, M., C. Simon, J. F. Aubry, A Fuzzy Probabilistic Approach for Determining Safety Integrity Level. IEEE Trans. on Fuzzy Syst. 2007, pp. 239-248.
  24. Volkanovski, A., M. Cepin, B. Mavko, Application of the Fault Tree Analysis for Assessment of Power System Reliability. Reliability Engineering & System Safety, Vol. 94, 2009, pp.1116-1127.

https://doi.org/10.24846/v19i4y201005

Designed by | ICI Bucharest
© Copyright 2024, All Rights Reserved