東華大學圖書館 |

A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis.

紀錄類型:	書目-電子資源 : Monograph/item
正題名/作者:	A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis./
作者:	Waghen, Kerelous Refaat Latef.
面頁冊數:	1 online resource (168 pages)
附註:	Source: Dissertations Abstracts International, Volume: 83-08, Section: B.
Contained By:	Dissertations Abstracts International83-08B.
標題:	Behavior. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28989914click for full text (PQDT)
ISBN:	9798759977568

A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis.
Waghen, Kerelous Refaat Latef.

A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis. - 1 online resource (168 pages)

Source: Dissertations Abstracts International, Volume: 83-08, Section: B.

Thesis (Ph.D.)--Ecole Polytechnique, Montreal (Canada), 2020.

Includes bibliographical references

The thesis develops a new methodology for diagnosis and prognosis of faults in a complex system, called Interpretable logic tree analysis (ILTA), which combines knowledge extraction techniques from knowledge discovery in databases (KDD) and the fault tree analysis (FTA). The methodology combined the advantages of the both techniques for understanding the problem of diagnosis and prognosis of faults. Although fault trees provide interpretable models for determining the possible causes of a fault, its use for fault diagnosis in an industrial system is limited, due to the need for expert knowledge to describe cause-and-effect relationships between internal system processes. However, it will be interesting to exploit the analytical power of fault trees but built from explicit and unbiased knowledge extracted directly from databases on the causality of faults. Therefore, the ILTA methodology works analogously to the logic of the fault tree analysis model (FTA) but with minimal involvement of experts. This modeling approach joins the logic of experts to represent the hierarchical structure of faults in a complex system. The ILTA methodology is applied to failure risk management by providing two interpretable advanced logic models: a multi-level tree (MILTA) and a multilevel tree over time (ITCA). The MILTA model is designed to accomplish the task of diagnosing failure in complex systems. It is able to decompose a complex defect and graphically model its causal structure in a tree on several levels. As a result, an expert is able to visualize the influence of hierarchical cause and effect relationships leading to the main failure. In addition, quantifying these causes by assigning probabilities helps to understand their contribution to the occurrence of system failure. The second model is a logical tree interpretable in time (ITCA), designed to perform the task of prognosis of failure in complex systems. Based on a distribution of data over time, the ITCA model captures the effect of the aging of the system through the evolution of the fault causation structure. Thus, it describes the causal changes resulting from deterioration and aging over the life of the system.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2023

Mode of access: World Wide Web

ISBN: 9798759977568Subjects--Topical Terms:

532476
Behavior.
Index Terms--Genre/Form:

542853
Electronic books.

A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis.
LDR:06051nmm a2200361K 4500 001 2353433
005 20230306113831.5
006 m o d
007 cr mn ---uuuuu
008 241011s2020 xx obm 000 0 eng d
020 $a 9798759977568
035 $a (MiAaPQ)AAI28989914
035 $a (MiAaPQ)EPMontreal4225
035 $a AAI28989914
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Waghen, Kerelous Refaat Latef. $3 3693778
245 1 2 $a A Novel Data-Driven Fault Tree Methodology for Fault Diagnosis and Prognosis.
264 0 $c 2020
300 $a 1 online resource (168 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 83-08, Section: B.
500 $a Advisor: Ouali, Mohamed-Salah.
502 $a Thesis (Ph.D.)--Ecole Polytechnique, Montreal (Canada), 2020.
504 $a Includes bibliographical references
520 $a The thesis develops a new methodology for diagnosis and prognosis of faults in a complex system, called Interpretable logic tree analysis (ILTA), which combines knowledge extraction techniques from knowledge discovery in databases (KDD) and the fault tree analysis (FTA). The methodology combined the advantages of the both techniques for understanding the problem of diagnosis and prognosis of faults. Although fault trees provide interpretable models for determining the possible causes of a fault, its use for fault diagnosis in an industrial system is limited, due to the need for expert knowledge to describe cause-and-effect relationships between internal system processes. However, it will be interesting to exploit the analytical power of fault trees but built from explicit and unbiased knowledge extracted directly from databases on the causality of faults. Therefore, the ILTA methodology works analogously to the logic of the fault tree analysis model (FTA) but with minimal involvement of experts. This modeling approach joins the logic of experts to represent the hierarchical structure of faults in a complex system. The ILTA methodology is applied to failure risk management by providing two interpretable advanced logic models: a multi-level tree (MILTA) and a multilevel tree over time (ITCA). The MILTA model is designed to accomplish the task of diagnosing failure in complex systems. It is able to decompose a complex defect and graphically model its causal structure in a tree on several levels. As a result, an expert is able to visualize the influence of hierarchical cause and effect relationships leading to the main failure. In addition, quantifying these causes by assigning probabilities helps to understand their contribution to the occurrence of system failure. The second model is a logical tree interpretable in time (ITCA), designed to perform the task of prognosis of failure in complex systems. Based on a distribution of data over time, the ITCA model captures the effect of the aging of the system through the evolution of the fault causation structure. Thus, it describes the causal changes resulting from deterioration and aging over the life of the system.
520 $a La these developpe une nouvelle methodologie de diagnostic et de pronostic de defauts dans un systeme complexe, nommee Interpretable logic tree analysis (ILTA), qui combine les techniques d'extraction de connaissances a partir des bases de donnees « knowledge discovery in database (KDD) » et l'analyse d'arbre de defaut « fault tree analysis (FTA) ». La methodologie capitalise les avantages des deux techniques pour apprehender la problematique de diagnostic et de pronostic de defauts. Bien que les arbres de defauts offrent des modeles interpretables pour determiner les causes possibles a l'origine d'un defaut, leur utilisation pour le diagnostic de defauts dans un systeme industriel est limitee, en raison de la necessite de faire appel a des connaissances expertes pour decrire les relations de cause-a-effet entre les processus internes du systeme. Cependant, il sera interessant d'exploiter la puissance d'analyse des arbres de defaut mais construit a partir des connaissances explicites et non biaisees extraites directement des bases de donnees sur la causalite des fautes. Par consequent, la methodologie ILTA fonctionne de maniere analogue a la logique du modele d'analyse d'arbre de defaut (FTA) mais avec une implication minimale des experts. Cette approche de modelisation doit rejoindre la logique des experts pour representer la structure hierarchique des defauts dans un systeme complexe. La methodologie ILTA est appliquee a la gestion des risques de defaillance en fournissant deux modeles d'arborescence avances interpretables a plusieurs niveaux (MILTA) et au cours du temps (ITCA). Le modele MILTA est concu pour accomplir la tache de diagnostic de defaillance dans les systemes complexes. Il est capable de decomposer un defaut complexe et de modeliser graphiquement sa structure de causalite dans un arbre a plusieurs niveaux. Par consequent, un expert est en mesure de visualiser l'influence des relations hierarchiques de cause a effet menant a la defaillance principale. De plus, quantifier ces causes en attribuant des probabilites aide a comprendre leur contribution dans l'occurrence de la defaillance du systeme. Le modele ITCA est concu pour realiser la tache de pronostic de defaillance dans les systemes complexes. Base sur une repartition des donnees au cours du temps, le modele ITCA capture l'effet du vieillissement du systeme a travers de l'evolution de la structure de causalite des fautes. Ainsi, il decrit les changements de causalite resultant de la deterioration et du vieillissement au cours de la vie du systeme.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2023
538 $a Mode of access: World Wide Web
650 4 $a Behavior. $3 532476
650 4 $a Manufacturing. $3 3389707
650 4 $a Failure. $3 3561225
650 4 $a Artificial intelligence. $3 516317
650 4 $a Signal processing. $3 533904
650 4 $a Data processing. $3 680224
650 4 $a Construction. $3 3561054
650 4 $a Fault diagnosis. $3 3555094
650 4 $a Logic. $3 529544
650 4 $a Sensors. $3 3549539
650 4 $a Neural networks. $3 677449
650 4 $a Decision making. $3 517204
650 4 $a Probability. $3 518898
650 4 $a Industrial engineering. $3 526216
655 7 $a Electronic books. $2 lcsh $3 542853
690 $a 0395
690 $a 0800
690 $a 0546
710 2 $a ProQuest Information and Learning Co. $3 783688
710 2 $a Ecole Polytechnique, Montreal (Canada). $3 1018606
773 0 $t Dissertations Abstracts International $g 83-08B.
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=28989914 $z click for full text (PQDT)