Subspace method aided data-driven design of fault detection and isolation systems

Ding, Steven X.; Zhang, Ping; Naik, Amol S.; Ding, E.L.; Huang, Biao

doi:10.1016/j.jprocont.2009.07.005

Cited by 293 publications

(155 citation statements)

References 20 publications

Supporting

Mentioning

155

Contrasting

Order By: Relevance

“…Remark 1: It is tedious but straightforward to show that (22) is equivalent to the output equation in classical parity space approaches [14]- [17], if it is also derived from the closed-loop observer form (7,8);…”

Section: And Collect the Lumped I/os In The Past Window Intomentioning

confidence: 99%

“…[8], [9], aims at overcoming the modeling stage by using system identification techniques, e.g. prediction error methods (PEM) [10] and subspace identification (SID) methods [11].…”

mentioning

confidence: 99%

“…In fact, the recent literature shows that the state space matrices, (A, B,C, D), are not really needed in designing an FDI, e.g. [8], [9]. Based on SID [11], [13], these approaches require computing the left null of the extended observability matrix, directly identified from measurement data, without realizing (A, B,C, D).…”

mentioning

confidence: 99%

See 2 more Smart Citations

Robust Fault Detection With Statistical Uncertainty in Identified Parameters

Dong

Verhaegen

Gustafsson

2012

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

Detection of faults that appear as additive unknown input signals to an unknown LTI discrete-time MIMO system is considered. State of the art methods consist of the following steps. First, either the state space model or certain projection matrices are identified from data. Then, a residual generator is formed based on these identified matrices, and this residual generator is used for online fault detection. Existing techniques do not allow for compensating for the identification uncertainty in the fault detection. This contribution explores a recent data-driven approach to fault detection. We show first that the identified parametric matrices in this method depend linearly on the noise contained in the identification data, and then that the on-line computed residual also depends linearly on the noise. This allows an analytic design of a robust fault detection scheme, that takes both the noise in the online measurements as well as the identification uncertainty into account. We illustrate the benefits of the new method on a model of aircraft dynamics extensively studied in literature.

show abstract

Section: And Collect the Lumped I/os In The Past Window Intomentioning

confidence: 99%

“…[8], [9], aims at overcoming the modeling stage by using system identification techniques, e.g. prediction error methods (PEM) [10] and subspace identification (SID) methods [11].…”

mentioning

confidence: 99%

See 1 more Smart Citation

Robust Fault Detection With Statistical Uncertainty in Identified Parameters

Dong

Verhaegen

Gustafsson

2012

IEEE Trans. Signal Process.

View full text Add to dashboard Cite

show abstract

“…The main aim of the paper is to formulate a modelfree approach to FTC that does not require any information about the plant in real-time. Unlike in the so-called subspace approach for FTC by Ding et al (2009), we do not focus on designing an FDI module that implicitly requires the knowledge of the plant parameters. On the contrary, our approach does not include any explicit FD module.…”

Section: Introductionmentioning

confidence: 99%

Model-free reconfiguration mechanism for fault tolerance

Jain¹,

Yamé²,

Sauter³

2012

International Journal of Applied Mathematics and Computer Science

View full text Add to dashboard Cite

The problem of fault tolerant control is studied from the behavioral point of view. In this mathematical framework, the concept of interconnection among the variables describing the system is a key point. The problem is that the behavior we intend to control is not known. Therefore, we are interested in designing a fault accommodation scheme for an unknown behavior through an appropriate behavioral interconnection. Here we deal simply with the trajectories that are generated by the system in real time. These trajectories determine the behavior of a system in various (faulty/healthy) modes. Based on the desired interconnected behavior, only the trajectories that obey certain laws are selected. These laws, representing the desired behavior, can indeed be achieved by a regular interconnection. Thus, when the trajectories do not belong to a certain desired behavior, it is considered to be due to the occurrence of a fault in the system. The vantage point is that the fault tolerant control problem now becomes completely a model-free scheme. Moreover, no explicit fault diagnosis module is required in our approach. The proposed fault tolerance mechanism is illustrated on an aircraft during the landing phase.

show abstract

“…For instance, the benchmark model in [Odgaard et al, 2009] contains lookup tables, which defines complicated nonlinear mappings, and can only be approximated by continuous functions. The difficulties in modeling have stimulated the development of data-driven FDI methods in literature [Dong et al, 2009, Qin and Li, 2001, Ding et al, 2009. Based on subspace identification methods [Verhaegen and Verdult, 2007], the data-driven PSA approaches in [Qin andLi, 2001, Ding et al, 2009] require computing the left null space of the identified range space of the extended observability matrix, directly identified from measurement data, without realizing the state-space matrices, e.g.…”

Section: Introductionmentioning

confidence: 99%

Data Driven Fault Detection and Isolation of a Wind Turbine Benchmark

Dong

Verhaegen

2011

IFAC Proceedings Volumes

View full text Add to dashboard Cite

This paper investigates data-driven fault detection and isolation (FDI) designs for a wind turbine benchmark problem. The benchmark is described by a SimuLink model, which contains nonlinear lookup tables and unknown wind disturbances. Based on classical FDI design methods, a linearization of the SimuLink model into a standard state-space form and describing the linearization errors as perturbations may be necessary. To avoid these difficult modeling procedures, this paper applies a data-driven design method, which produces FDI filters directly based on the simulated data from the benchmark SimuLink model. The fixed-value sensor faults therein are especially targeted. Moreover, we develop in this paper a new data-driven fault isolation scheme, via exploiting hardware redundancy in a plant. Based on this, a bank of robust data-driven detection filters are designed for the benchmark and implemented in parallel. The simulation results show the effectiveness of the applied data-driven scheme.

show abstract

Subspace method aided data-driven design of fault detection and isolation systems

Cited by 293 publications

References 20 publications

Robust Fault Detection With Statistical Uncertainty in Identified Parameters

Robust Fault Detection With Statistical Uncertainty in Identified Parameters

Model-free reconfiguration mechanism for fault tolerance

Data Driven Fault Detection and Isolation of a Wind Turbine Benchmark

Contact Info

Product

Resources

About