Fault detection and Isolation from an identified MIMO Takagi-Sugeno model of a bioreacto

Chenikher, Salah; Cassar, Jean-Philippe; Pékpé, Komi Midzodzi

doi:10.3182/20090630-4-es-2003.00207

Cited by 6 publications

(6 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Input-output relation is first established on a given time-window and a projection matrix is then computed with the objective to eliminate not only the unknown state but also the model parameters ( [13]). …”

Section: Methodsmentioning

confidence: 99%

“…The projection method needs only the input-output measurements and the weighting functions Y k and Φ j k from the matrix relation given by equation (13) and does not take account of the system parametersH j .In the following, we demonstrate the insensitivity of residuals to constant parameters uncertainties in a one local model. The generalized form described the TS model and expressed by equation (13) is given in appendix B, equation (39).…”

Section: Sensitivity Of Residuals To Model Uncertaintiesmentioning

confidence: 98%

“…This method needs to know the local models parameters and the weighting function of each mode. -The second method is based on the projection of an output matrix into the right kernel of a weighted Hankel matrix ( [13]). This method needs to know the weighting function of each mode and does not need the parameters values of the TS models.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Robust mode recognition in Hybrid Dynamical Systems with nonlinear modes

Zouari

Pékpé

Cocquempot

et al. 2011

2011 Chinese Control and Decision Conference (CCDC)

View full text Add to dashboard Cite

Switching detection and active mode recognition in Hybrid Dynamical Systems (HDS) with nonlinear continuous dynamics are considered in this paper. The nonlinear modes are represented by Takagi-Sugeno models. Two methods based on projection techniques are proposed to generate a bank of residuals which are mode indicators: the first one is the well-known parity space method which uses the TS model parameters, the second one is a data-based projection method that does not need the parameter values of the TS model. Robustness of the two methods against parameter uncertainties is discussed. An illustrative example is provided to show the effectiveness of the proposed methods.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Sensitivity Of Residuals To Model Uncertaintiesmentioning

confidence: 98%

See 1 more Smart Citation

Robust mode recognition in Hybrid Dynamical Systems with nonlinear modes

Zouari

Pékpé

Cocquempot

et al. 2011

2011 Chinese Control and Decision Conference (CCDC)

View full text Add to dashboard Cite

show abstract

“…×L is a matrix, where the number of independent rows is equal or less than the number of independent columns, then the right kernel of Γ is given by: (20) where Π Γ is the right projection matrix of Γ, and consequently we have:…”

Section: Propositionmentioning

confidence: 99%

“…If the fault fluctuates instantaneously and H i V i = 0, it is the same case as for constant bias sensor faults. But if the fault fluctuates instantaneously and H i V i = 0, a more convenient decision algorithm should be developed and this case is not treated in this paper for brevity reasons (for signature table the reader is referred to [20] and [21]). …”

Section: A Sensor Fault Isolabilitymentioning

confidence: 99%

A parameter-free method for sensor fault detection and isolation in bilinear systems

Hakem

Pékpé

Cocquempot

2012

Proceedings of 2012 UKACC International Conference on Control

View full text Add to dashboard Cite

This paper is concerned with Fault Detection and Isolation (FDI) and more specifically it focuses on a parameterfree residual generation method. The residual signals are obtained by projecting the measured signals onto the kernel of an extended input matrix, which depends on the structure of the system model. The method was not easily applicable in real-world applications due to a high computational complexity. In that paper, fault indicators are constructed differently, using kernels properties, to avoid this complexity problem. A simulated electromechanical actuator example is taken to illustrate the applicability of the method.

show abstract