A fault modeling approach for Input/Output Automata

Nke, Yannick; Lunze, Jan

doi:10.3182/20110828-6-it-1002.00624

Cited by 5 publications

(2 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…All faults f belong to a known fault set F and all corresponding models from the model set {A f , (f ∈ F)} of the faulty plant are assumed to be known. The models A f , (f ∈ F) of the faulty plant can be assumed from the model A 0 of the faultless plant by applying the error relations described in [5] to it. If the faultf ∈ F is present, the plant is modeled by P Af .…”

Section: Plant Modelsmentioning

confidence: 99%

“…According to (9), each pair of states from the state set Z has to be considered, when a controller C is defined for an automaton A. From the construction of the common model A ∩ in (17) based on the intersection operation in Definition 4.2 and the assumption in (5), it follows that the state set Z ∩ of the common model A ∩ is of size |Z 0 |. That is, the complexity of the reconfiguration step is quadratic in the number of plant states such that it is reasonable to apply the method to real-world systems.…”

Section: Proofmentioning

confidence: 99%

See 1 more Smart Citation

Fault-tolerant control of deterministic I/O automata with ambiguous diagnostic result

Schuh

Lunze

2016

2016 13th International Workshop on Discrete Event Systems (WODES)

View full text Add to dashboard Cite

A method for the active fault-tolerant control of discrete event systems modeled by deterministic Input/Output automata is proposed that can be applied if the diagnostic result consists of a set of fault candidates rather than a single fault. Based on the ambiguous diagnostic result a common model of the faulty plant is constructed that describes the identical part of the behavior of the plant under all fault candidates. Then the reconfiguration of the controller is performed based on this common model. It is proved that the resulting reconfigured tracking controller steers the faulty plant into the desired final state, no matter which fault actually occurred.

show abstract

Section: Plant Modelsmentioning

confidence: 99%

Section: Proofmentioning

confidence: 99%