Retour au site du LAAS-CNRS

Laboratoire d’analyse et d’architecture des systèmes

Publications de l'équipe DISCO

Choisir la langue : FR | EN

876documents trouvés

18164
31/08/2018

Chronicle Discovery for Diagnosis from Raw Data: A Clustering Approach

A.SAHUGUEDE, E.LE CORRONC, M.V.LE LANN

DISCO

Manifestation avec acte : IFAC International Symposium on Fault Detection Supervision and Safety of Technical Processes ( SAFEPROCESS ) 2018 du 29 août au 31 août 2018, Varsovie (Pologne), Août 2018, 8p. , N° 18164

Lien : https://hal.laas.fr/hal-01817529

Diffusable

Plus d'informations

Abstract

Chronicles are temporal patterns well suited for an abstract representation of the behavior of dynamic systems. For fault diagnosis, chronicles describe the nominal and faulty behaviors of the process. Powerful algorithms allow the recognition of chronicles in the flow of observations of the system and appropriate actions can be taken when a faulty situation is recognized. However, designing chronicles is not a trivial thing to do. The increasing complexity and capacity of data generation of highly-advanced processes cause the acquisition of a complete model difficult. This paper focuses on the problem of discovering chronicles that are representative of a system behavior from direct observations. A clustering approach to this problem is considered. The chronicle discovery algorithm proposed here designs chronicles with minimal knowledge of the system to diagnose. Furthermore, unprocessed data obtained directly from the system can be used in this clustering algorithm. Finally, the chronicle discovery algorithm proposed in this paper is illustrated on a sport performance monitoring device for a diagnosis of movement deviations in the temporal domain, in the event domain, or both, considered as faults for the athlete.

143879
18403
31/08/2018

Decentralized Diagnosis via Structural Analysis and Integer Programming

C.PEREZ, E.CHANTHERY, L.TRAVE-MASSUYES, J.SOTOMAYOR, C.ARTIGUES

DISCO, PUCP, ROC

Manifestation avec acte : IFAC International Symposium on Fault Detection Supervision and Safety of Technical Processes ( SAFEPROCESS ) 2018 du 29 août au 31 août 2018, Varsovie (Pologne), Août 2018, 7p. , N° 18403

Lien : https://hal.archives-ouvertes.fr/hal-01882324

Diffusable

Plus d'informations

Abstract

Centralized fault diagnosis architectures are sometimes prohibitive for large-scale interconnected systems such as distribution systems, telecommunication networks, water distribution networks, fluid power systems. This paper presents a decentralized fault diagnosis method for continuous systems that only requires the knowledge of local models and limited knowledge of their neighboring subsystems. The notion of Fault-Driven Minimal Structurally Overdetermined (FMSO) set is used as the corner stone of the design of residual generators for the design of decentralized fault diagnosis for systems that have constraints of confidentiality, distance or limited access to some information. Binary integer linear programming (BILP) is used to optimize the choice of FMSO sets in each local subsystem.

145458
18243
30/08/2018

Computer-aided Diagnosis via Hierarchical Density Based Clustering

T;OBRY, L.TRAVE-MASSUYES, A.SUBIAS

DISCO

Manifestation avec acte : International Workshop on Principles of Diagnosis ( DX ) 2018 du 27 août au 30 août 2018, Varsovie (Pologne), Août 2018, 8p. , N° 18243

Lien : https://hal.laas.fr/hal-01847563

Diffusable

Plus d'informations

Abstract

When applying non-supervised clustering, the concepts discovered by the clustering algorithm hardly match business concepts. Hierarchical clustering then proves to be a useful tool to exhibit sets of clusters according to a hierarchy. Data can be analyzed in layers and the user has a full spectrum of clusterings to which he can give meaning. This paper presents a new hierarchical density-based algorithm that advantageously works from compacted data. The algorithm is applied to the monitoring of a process benchmark, illustrating its value in identifying different types of situations , from normal to highly critical.

144398
18404
30/08/2018

Optimal test / sensor selection problems formalized as integer programs

C.ARTIGUES, BBASSENE, E.CHANTHERY, A.GASMI, L.TRAVE-MASSUYES

ROC, DISCO

Manifestation avec acte : International Workshop on Principles of Diagnosis ( DX ) 2018 du 27 août au 30 août 2018, Varsovie (Pologne), Août 2018, 8p. , N° 18404

Lien : https://hal.archives-ouvertes.fr/hal-01882322

Diffusable

Plus d'informations

Abstract

Diagnosis is the reasoning leading to the identification of the cause of a problem. Given a system instrumented with a set of sensors, diagnosis can be performed thanks to diagnosis tests that are designed from the system model. The tests only involve measured variables and can be checked with the measured values. The configuration of tests that pass and tests that do not pass provides a way to isolate the faults. However the fact that all faults are discriminable, i.e. the system is fully diagnosable, depends on the set of tests and hence of the sensors that are placed on the system. The number of tests that can be designed is generally huge and more than sufficient to achieve full or maximal diagnosability. This paper addresses several variants of the problem of selecting the set of tests/sensors so that diagnosability is maximal and a cost criterion is minimized. The variant problems are formalized in the integer programming framework.

145460
18191
24/08/2018

Mapping Chronicles to a k-dimensional Euclidean Space via Random Projections

A.SAHUGUEDE, S.FERGANI, E.LE CORRONC, M.V.LE LANN

DISCO

Manifestation avec acte : IEEE International Conference on Automation Science and Engineering ( CASE ) 2018 du 20 août au 24 août 2018, Munich (Allemagne), Août 2018, 6p. , N° 18191

Lien : https://hal.laas.fr/hal-01817539

Diffusable

Plus d'informations

Abstract

This paper is concerned with an innovative strategy that maps chronicles, that are timed discrete event models, to a k-dimensional Euclidean space via random projections. The proposed approach is a projection that takes into account both characteristics of events, namely event types, and temporal constraints of chronicles. This will lead to an unbounded convex polytope in the Euclidean space that contains all the possible instances of the corresponding chronicle. It allows to easily and efficiently compare chronicles. Such comparisons are useful in a fault diagnosis purpose to discriminate chronicles representing behaviors of dynamic processes. Examples and preliminary results are provided in this paper to introduce the proposed methodology.

144035
18200
11/07/2018

Vers des systèmes plus autonomes : contributions autour de la tâche de diagnostic dans une architecture embarquée

E.CHANTHERY

DISCO

Habilitation à diriger des recherches : 11 Juillet 2018, 155p., Président: J.ZAYTOON, Rapporteurs: V.COCQUEMPOT, P.DAGUE, D.LEFEBVRE, Examinateurs: A.SUBIAS, Référente: L.TRAVE-MASSUYES , N° 18200

Lien : https://hal.archives-ouvertes.fr/tel-01882329

Diffusable

Plus d'informations

Abstract

In my presentation, I will first review my teaching activities at INSA Toulouse, as well as my various responsibilities. In a second step, I will give the major axes of my research whose objective is to increase the autonomy of systems by designing appropriate and effective fault diagnosis modules. The detection and isolation of faults help the system to react correctly to the different situations it has to face, thus contributing greatly to its autonomy. I will focus on two major points. Firstly, the link between diagnosis and prognosis will be developed in the context of hybrid dynamic systems. The prognosis aims at predicting the future states of the system. A health management architecture will be presented and then a common modeling framework. I will show that this work has made it possible to standardize the joint formalisms of diagnosis and prognosis. A diagnostic algorithm whose results can be interpreted by the prognostic module will be presented. This algorithmic work, followed by an application in the real case of a rover, constitutes a major contribution for the intrinsic coupling of diagnostic and prognostic algorithms. The second point will be distributed and decentralized diagnosis for continuous dynamics systems. Structural analysis will be proposed as a solution for test generation in complex systems. Despite its apparent simplicity, it provides a powerful set of tools for analyzing and inferring information about the system. Moreover, it has the advantage of being applied indifferently to linear and non-linear systems. The presentation will show how the notions of diagnosis have been adapted in the decentralized and distributed framework, up to the formulation of an optimization problem linked to the choice of a subset of subsystem-level diagnostic tests. . Finally, the last part of my presentation will focus on my research project and present my perspectives for future years.

Résumé

Dans ma présentation d’HDR, j’exposerai dans un premier temps mes activités d’enseignement à l’INSA de Toulouse, ainsi que mes diverses responsabilités et investissements. Dans un deuxième temps, je donnerai les grands axes de ma recherche dont l’objectif est d’accroitre l’autonomie des systèmes en concevant des modules de diagnostic de fautes efficaces et adaptés. La détection et l’isolation des fautes qui peuvent survenir permettent en effet au système de réagir correctement aux différentes situations dans lesquelles il se trouve, contribuant ainsi grandement à son autonomie. Je me focaliserai sur deux points particuliers. Tout d’abord le lien entre le diagnostic et le pronostic sera développé dans le cadre des systèmes à dynamiques hybrides, le pronostic consistant à prédire les états futurs du système. Une architecture de gestion de santé sera présentée puis un cadre commun de modélisation. Je montrerai que ce travail a permis d’uniformiser les formalismes conjoints de diagnostic et de pronostic. Un algorithme de diagnostic dont les résultats sont interprétables par le module de pronostic sera présenté. Ce travail algorithmique, suivi d'une application dans un cas réel de rover, constitue une contribution majeure pour le couplage intrinsèque d’algorithmes de diagnostic et de pronostic. Le deuxième point portera sur le diagnostic décentralisé et distribué pour des systèmes à dynamique continue. L’analyse structurelle sera proposée comme solution pour la génération de tests dans le cadre des systèmes complexes. Malgré son apparente simplicité, cette dernière fournit un ensemble d'outils puissants pour analyser et inférer des informations sur le système. Par ailleurs, elle a l'avantage de s'appliquer indifféremment aux systèmes linéaires et non linéaires. L’exposé montrera comment les notions de diagnostic ont été adaptées dans le cadre décentralisé et distribué, jusqu’à la formulation d’un problème d'optimisation lié au choix d'un sous-ensemble de tests de diagnostic au niveau des sous-systèmes. Enfin, la dernière partie de mon exposé portera sur mon projet de recherche et présentera mes perspectives pour les années futures.

Mots-Clés / Keywords
Diagnostic; Pronostic; Systèmes autonomes; Approche décentralisée; Approche distribuée; Sélection de tests; Autonomous systems; Decentralized approach; Distributed approach; Test selection;

144117
18153
20/06/2018

A Study of Evacuation Planning for Wildfires

C.ARTIGUES, E.HEBRARD, Y.PENCOLE, A.SCHUTT, P.STUCKEY

ROC, DISCO, CSIRO, Univ of Melbourne

Rapport LAAS N°18153, Juin 2018, 17p.

Lien : https://hal.archives-ouvertes.fr/hal-01814083

Diffusable

Plus d'informations

Abstract

The GEO-SAFE project gathers researchers and fire emergency practitioners from EU and Australia with the aim to design innovative models and efficient response tools based on optimization methods for fighting wildfires. In this paper, we consider an evacuation planning problem issued from discussions with practitioners, where a wildfire is threatening a region with intermediate populated centres. As in earlier approaches in case of flood, we use a constraint optimization model involving malleable tasks to represent the evacuation of a population and a cumulative constraint per route segments. Indeed, in order to mitigate congestion risks, the authorities may delay the start of the evacuation but they may also affect the rate of evacuation by modulating the method used to raise the alarm. However, we consider a different objective: we maximize the minimum " safety margin " , weighted by the population, over every road segment. We introduce a new heuristic and a global flow constraint propa-gator. Moreover, we also propose an instance generator based on a random generation of road networks and basic fire propagation models. This generator produces challenging benchmarks even with very few evacuation tasks. Finally, we report the results of extensive computational experiments done using CP Optimizer.

143836
18154
20/06/2018

GeoSafe – Evacuation planning problems

C.ARTIGUES, E.HEBRARD, Y.PENCOLE, A.SCHUTT, P.STUCKEY

ROC, DISCO, CSIRO, Univ of Melbourne

Rapport LAAS N°18154, Juin 2018, 10p.

Lien : https://hal.archives-ouvertes.fr/hal-01814177

Diffusable

Plus d'informations

Abstract

This paper aims at identifying new challenging variants of the evacuation planning problems, especially in case of bush fires, based on exchanges with practitioners in the context of the GeoSafe project. A large amount of work has been carried out at NICTA mainly in the context of floodings, which can be easily transposed to evacuation in case of fires. .We present a state-of-the-art review of these papers and we provide new research directions for the GeoSafe project.

143837
18155
20/06/2018

Data Instance generator and optimization models for evacuation planning in the event of wildfire

C.ARTIGUES, E.HEBRARD, Y.PENCOLE, A.SCHUTT, P.STUCKEY

ROC, DISCO, CSIRO, Univ of Melbourne

Rapport LAAS N°18155, Juin 2018, 18p.

Lien : https://hal.archives-ouvertes.fr/hal-01814063

Diffusable

Plus d'informations

Abstract

One critical part of decision support during the response phase to a wildfire is the ability to perform large-scale evacuation planning. While in practice most evacuation planning is principally designed by experts using simple heuristic approaches or scenario simulations, more recently optimization approaches to evacuation planning have been carried out, notably in the context of floodings. Evacuation planning in case of wildfires is much harder as wildfire propagations are inherently less predictable than floods. This paper present a new optimization model for evacuation planning in the event of wildfire aiming at maximizing the temporal safety margin between the evac-uees and the actual or potential wildfire front. As a first contribution, an open-source data instance generator based on road network generation via quadtrees and a basic fire propagation model is proposed to the community. As a second contribution we propose 0–1 integer programming and constraint programming formulations enhanced with a 1 simple compression heuristic that are compared on 240 problem instances build by the generator. The results show that the generated instances are computationally challenging and that the contraint programming framework obtains the best performance.

143838
18401
15/06/2018

Diagnostic préférentiel à base de modèles discrets pour des fautes intermittentes et permanentes

V.BOUZIAT, X.PUCEL, S.ROUSSEL, L.TRAVE-MASSUYES

ONERA, DISCO

Manifestation avec acte : Journées d'Intelligence Artificielle Fondamentale ( JIAF ) 2018 du 13 juin au 15 juin 2018, Amiens (France), Juin 2018, 8p. , N° 18401

Lien : https://hal.archives-ouvertes.fr/hal-01928837

Diffusable

Plus d'informations

Résumé

Nous nous intéressons dans cet article au diagnostic de pannes intermittentes et permanentes dans des systèmes à événements discrets. Nous proposons un formalisme de modélisation logique associé à des préférences conditionnelles destiné à produire un diagnostic unique à chaque pas de temps. Cette approche est susceptible de provoquer des interblocages entre le système surveillé et son diagnostiqueur. L'objet de cet article est de détecter ces interblocages dès la conception, par une méthode basée sur le model-checking.

145454
Les informations recueillies font l’objet d’un traitement informatique destiné à des statistiques d'utilisation du formulaire de recherche dans la base de données des publications scientifiques. Les destinataires des données sont : le service de documentation du LAAS.Conformément à la loi « informatique et libertés » du 6 janvier 1978 modifiée en 2004, vous bénéficiez d’un droit d’accès et de rectification aux informations qui vous concernent, que vous pouvez exercer en vous adressant à
Pour recevoir une copie des documents, contacter doc@laas.fr en mentionnant le n° de rapport LAAS et votre adresse postale. Signalez tout problème de dysfonctionnement à sysadmin@laas.fr. http://www.laas.fr/pulman/pulman-isens/web/app.php/