Wednesday , June 20 2018

State Estimation via Observers with Unknown Inputs: Application to a Particular Class of Uncertain Takagi-Sugeno Systems

Wafa JAMEL1, Atef KHEDHER2, Nasreddine BOUGUILA1, Kamel Ben OTHMAN2
1 ATSI, National Engineering School of Monastir
Rue Ibn El Jazzar, 5019 Monastir-Tunisia
(wafa_jamel, nasreddinebouguila)@yahoo.fr
2 LARA Automatique, National Engineering School of Tunis
BP 37, le Belvédère, 1002 Tunis
(khedher_atef, kamelbenothman)@yahoo.fr

Abstract: This paper deals with the design of a multiple observer allowing estimating the state vector of a nonlinear system described by a Takagi-Sugeno multiple model subject to modelling and input uncertainties which are considered as unknown inputs. The main contribution of the paper is the conception of a multiple observer based on the elimination of these unknown inputs. Convergence conditions are established in order to guarantee the convergence of the state estimation error. These conditions are expressed in Linear Matrix Inequality (LMI) formulation. An example of simulation is given to illustrate the proposed method.

Keywords: Multiple model approach, Takagi-Sugeno models, multiple observer, unknown inputs, state estimate, modelling and input uncertainties.

>>Full text
CITE THIS PAPER AS:
Wafa JAMEL, Atef KHEDHER, Nasreddine BOUGUILA, Kamel Ben OTHMAN, State Estimation via Observers with Unknown Inputs: Application to a Particular Class of Uncertain Takagi-Sugeno Systems, Studies in Informatics and Control, ISSN 1220-1766, vol. 19 (3), pp. 219-228, 2010.

1. Introduction

State estimation plays a significant role in the context of monitoring and/or of diagnosis of systems. It is an analytical source of redundancy used to generate failure symptoms of the system by making a comparison between the real behaviour signals of the system and the estimated signals. A non desired variation between these signals indicates the possible presence of faults affecting the system. These faults indicators are named residues. Their generation is based on the use of the state observers.

A state observer is a dynamical system allowing the state reconstruction from the system model and the measurements of its inputs and outputs [20]. In fact, the observer controlled by the same inputs applied to the system is able to provide the same output signals provided that the model employed reproduces with precision the behaviour of the system to be supervised. The observer design can be delicate according to the type and the complexity of the considered model.

Two types of models are distinguished according to the linear or non linear character of the system. Linear models have simple structures. They are the base of several applications and research works.

In the case of linear systems, observers can be designed for uncertain systems with time-delay perturbations [8] and unknown input systems [7]. Several researches were achieved concerning the state estimation in the presence of both known and unknown inputs [26], [29], [7]. These works can be gathered into two categories. The first one supposes an a priori knowledge of information on these non measurable inputs; in particular, Johnson [17] proposes a polynomial approach and Meditch [21] suggests approximating the unknown inputs by the response of a known dynamic system. The second category proceeds either by estimation of the unknown inputs [18], [19], or by their complete elimination from the system equations [11].

However, in the majority of real cases the nonlinear nature of the process cannot be neglected. The assumption of linearity is checked only locally around an operating point. Real physical processes present complex behaviours with nonlinear laws. As, it is delicate to synthesize an observer for a nonlinear system, the multiple model approach constitutes a tool which is largely used in the modelling of nonlinear systems [22], [6].

The principle of the multiple model approach is based on the reduction of the system complexity by the decomposition of its operating space in a finite number of operating zones. Each zone is characterized by a local model (named also, sub-model). Each sub-model is a simple and linear system around an operating point. The relative contribution of each sub-model is quantified with the help of a weighting function. The global behaviour of the nonlinear system is obtained by the sum of the local models balanced by the weighting functions.

Various studies dealing with the presence of unknown inputs acting on the nonlinear system were published [3], [4], [18], [19]. The problem of state estimation of nonlinear systems submitted to uncertainties has received considerable attention [13], [1], [24], [28]. In practice, there are many situations where some of the system inputs are inaccessible. The recourse to the use of an unknown input observer is then necessary in order to be able to estimate the state of the considered system. For state estimation, the suggested technique consists in associating to each local model a local unknown input observer. The multiple observer or global observer is the sum of the local observers weighted by the weighting functions associated to the local models [25].

In this paper, the problem of state estimation of an uncertain Takagi-Sugeno multiple model is addressed. The purpose of this work is to extend the principle of the design of observers with unknown inputs to uncertain system case. Only model’s and input uncertainties are considered in this paper.

Others works dealing with uncertain systems choose to estimate the sate using different kinds of observers, such as sliding mode observer [1], Proportional and Proportional Integral observer [23]. The main contribution in this paper is the development of an unknown input multiple observer for uncertain nonlinear systems modelled by Takagi-Sugeno models. The convergence conditions of the sate estimation error are expressed in terms of linear matrix inequalities (LMI).

The paper is organized as follows. Section 2 recalls the multiple model approach. In section 3, the multiple observer of a system with unknown inputs is presented.

Section 4 presents the main results concerning the synthesis of a multiple observer to estimate the state of nonlinear system submitted to modelling and inputs uncertainties. Finally, in section 5, a numerical example is given to show the validity of the proposed methodology.

REFERENCES

  1. Akhenak, A., M. Chadli, J. Ragot, D. Maquin, Design of Sliding Mode Unknown Input Observer for Uncertain Takagi-Sugeno Model, 15th Mediterranean Conference on Control and Automation, MED’07, Athens, Greece, June 27-29, 2007.
  2. Akhenak, A., D. Maquin, J. Ragot, Process Input Estimation with a Multimodel. Application to Communication, Mediterranean Conference on Control and Automation, Limassol, Cyprus, June 27-29, 2005.
  3. Akhenak, A., Conception d’observateurs non linéaires par approche multimodèle: application au diagnostic. Thèse de doctorat, Institut National Polytechnique de Lorraine, December 16, 2004.
  4. Akhenak, A., M. Chadli, J. Ragot, D. Maquin, Design of Robust Observer for Uncertain Takagi-Sugeno Models, IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2004, Budapest, Hungary, 25-29 July, 2004.
  5. Angelov, P., D. Filev, An Approach to Online Identification of Takagi-Sugeno Fuzzy Models, IEEE Transactions on systems, Man and Cybernetics, Part B, Vol. 34, Nr. 1, 2004, pp. 484-498.
  6. Chadli, M., D. Maquin, J. Ragot, Multiple Observers for Discrete-time Multiple Models, 5th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes, Safeprocess’2003, Washington, D.C., USA, June 9-11, 2003.
  7. Darouach, M., M. Zasadzinski, S. J. SXu, Full-order Observers for Linear Systems with Unknown Inputs, IEEE Transactions on Automatic Control, 39(3), 1994, pp. 606-609.
  8. Fan, K. K., J. G. Hsieh, LMI Approach to Design of Robust State Observer for Uncertain Systems with Time-delay Perturbation, IEEE ICIT’02, Bangkok, Thailand, 2002, pp. 1111-1115.
  9. Filev, D., Fuzzy Modeling of Complex Systems, International Journal of Approximate Reasoning, 5(3), 1991, pp. 281-290.
  10. Gasso, K., G. Mourot, J. Ragot, Structure Identification in Multiple Model Representation: Elimination and Merging of Local Models, IEEE Conference on Decision and Control, 3, 2001, pp. 2992-2997.
  11. Guan, Y., M. Saif, A Novel Approach to the Design of Unknown Input Observers, IEEE Transactions on Automatic Control, Vol. 36, Nr. 5, 1991, pp. 632-635.
  12. Ichalal, D., B. Marx, J. Ragot, D. Maquin, Simultaneous State and Unknown Inputs Estimation with PI and PMI Observers for Takagi Sugeno Model with Unmeasurable Premise Variables, 17th Mediterranean Conference on Control and Automation, MED’09, Thessaloniki, Greece, June 24-26, 2009.
  13. JAMEL, W., A. KHEDHER, N. BOUGUILA K. Benothman, Design of Multiple Observers for a Class of Uncertain Takagi-Sugeno Multiple Models, 6th International Conference on Electrical Systems and Automatic Control -JTEA 10, Hammamet, Tunisia, March 26-28, 2010.
  14. Johansen, T. A., R. Babuska, Multiobjective Identification of Takagi-Sugeno Fuzzy Models, IEEE Transactions on Fuzzy Systems, Vol. 11, Nr. 6, 2003, pp. 847-860.
  15. Johansen, T. A., A .B. Foss, Constructing NARMAX using ARMAX Models, International Journal of Control, 58(5), 1993, pp. 1125-1153.
  16. JOHANSEN, T. A., A. B. Foss, Nonlinear Local Model Representation for Adaptive Systems, Singapore International Conference on Intelligent Control and Instrumentation, Singapore, February 17-21, 1992.
  17. JOHNSON, C. D., Observers for Linear Systems with Unknown and Inaccessible Inputs, International Journal of Control, 21, 1975, pp. 825-831.
  18. Khedher, A., K. Benothman, D. Maquin, M. Benrejeb, State and Sensor Faults Estimation via a Proportional Integral Observer, 6th international multi-conference on Systems signals & devices SSD’09, Djerba, Tunisia, March 23-26, 2009.
  19. Khedher, A., K. Benothman, Proportional Integral Observer Design for State and Faults Estimation: Application to the Three Tanks System, International review of automatic control. 3(2), 2010, pp. 115-124.
  20. Luenberger, D. G., An Introduction to Observers, IEEE Transactions on Automatic Control, Vol. 16, Nr. 6, 1971, pp. 596-602.
  21. MEDITCH, J. S., G. H. HOSTETTER, Observers for Systems with Unknown and Inaccessible Inputs, International Journal of Control, 19, 1971, pp. 637-640.
  22. Murray-Smith, R., T. A. Johansen, Multiple Model Approaches to Modelling and Control, Taylor and Francis, London, 1997.
  23. Orjuela, R., Contribution à l’estimation d’état et au diagnostic des systèmes représentés par des multimodèles. Thèse de doctorat, Institut National Polytechnique de Lorraine, 2008.
  24. Park, J. H., G. T. Park, Adaptive Fuzzy Observer with Minimal Dynamic Order for Uncertain Nonlinear Systems, IEE Proceedings Control Theory and Applications, 150(2), 2003, pp. 189-197.
  25. Patton, R. J., J. Chen, Observer-based Fault Detection and Isolation Robustness and Applications, Control Engineering Practice, 5, Nr. 5, 1997, pp. 671-682.
  26. Sename, O., Unknown Input Robust Observer for Time Delay System, IEEE Conference on Decision and Control, 2, 1997, pp. 1629-1630.
  27. TAKAGI, T., M. SUGENO, Fuzzy Identification of Systems and its Application to Modelling and Control, IEEE Transaction Systems, Man and Cybernetics, 15(1), 1985, pp. 116-132.
  28. Tong, S., H. H. Li, Observer-based Robust Fuzzy Control of Nonlinear Systems with Parametric Uncertainties, Fuzzy Sets and Systems, 131(2), 2002, pp. 165-184.
  29. Wang, S. H., E. J. Davison, P. Dorato, Observing the States of Systems with Unmeasurable Disturbances, IEEE Transactions on Automatic Control, 20, 1995, pp. 716-717.
  30. Yang, F. R. W. Wilde, Observers for Linear Systems with Unknown Inputs, IEEE Transactions on Automatic Control, 33, 1988, pp. 677-681.

https://doi.org/10.24846/v19i3y201002