Search

International audience; The safety control process of industrial systems (considered to be dynamic systems) needs to take in account physical processes (e.g. building ageing), informational processes (data collection and processing), decisional processes (data aggregation), and has to consider various constraints (e.g. economic and regulatory). The improvement of informational and decisional processes with the aim of controlling physical processes is based on the development of models and algorithms for measurement, assessment, control, diagnosis and prognostic. In the domain of dam management, assessment of reliability and safety, fault diagnosis, and corrective action proposals are carried out by expert engineers during dam reviews. With the perspective to assist these expert engineers, it is of great importance to develop methods and tools to manage the dynamic behaviour of dams and to model the processes at the same level of abstraction that is used by experts. In this chapter, the authors tackle the cognitive process of the diagnosis by means of a formal multi-modelling method and a diagnosis algorithm. The multi-modelling method called Timed Observations Method for Diagnosis (TOM4D) is based on the elaboration of four models: a Structural Model describing the relations between the components of the system, a Functional Model providing the relations between the values of the process variables (i.e. a set of mathematical functions), a Behavioural Model defining the states of the process and the discrete events firing the state transitions, and a Perception Model composed of a set of abstract variables, a set of thresholds associated to these variables and a set of constraints. The resulting process allows the automatic fault detection, identification and diagnosis and it is applied to hydraulic dam safety.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]RIVAGE; International audience; This paper proposes the use of the Timed Observation theory as a powerful framework for model-based diagnosis. In fact, this theory provides a global formalism for modelling a dynamic system (TOM4D), for characterizing and computing diagnoses of the system under investigation.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]RIVAGE; International audience; It is now well-known that the size of the model is the bottleneck when using model-based approaches to diagnose complex systems. To answer this problem, decompositional and multi modelling approaches have been proposed. In this paper, we propose a multi-modelling method called TOM4D (Timed Observations Modelling for Diagnosis) able to cope with dynamic aspects. It relies on four models: perception, structural, functional and behaviour models. The behaviour model is described through system component models as a set of component behaviour models and the global diagnosis is computed from the component diagnoses (also called local diagnoses). Another problem, which is far less considered, is the size of the diagnosis itself. However, it can also be huge enough, especially when dealing with dynamic system. To solve this problem, we propose in this paper to use The Timed Observation Theory. In this context, we characterize the diagnosis using TOM4D and the timed observation theory. We show their relevance to get a tractable representation of diagnosis. To illustrate the impact on the diagnosis size, experimental results on a hydraulic example are given.

Dams are heterogeneous structures featured by complex behaviours that evolve through time due to natural ageing. As a consequence, it is essential to develop modelling approaches for diagnosis taking into account the temporal aspect to comply the assessment of reliability and safety at the current time, to diagnose the causes of reliability and safety deterioration in the past and to forecast dam reliability and safety at different time scales. In this paper, we propose a multi-modelling method called TOM4D (Timed Observations Modelling for Diagnosis) able to cope with dynamic aspects. It relies on four models: perception, structural, functional and behavioural models. We use this methodology to assess the dam behaviour at different times and the detection of behaviour deterioration. The resulting model allows to characterize and to compute diagnosis of the system under investigation. The diagnosis was applied at two levels, the component level (local diagnosis) and the dam level (global diagnosis). Only a few process variables are used to model the dam behaviour and its components. These process variables are linked to real, visual and monitoring measurements carried out on the dam (piezometry, outflows, detection of leakage...). An application on a real case is performed., Les barrages sont caractérisés par des comportements complexes qui évoluent au cours du temps du fait d’un vieillissement naturel. Il est par conséquent pertinent de développer des approches de modélisation prenant en compte les aspects temporels. Le but de cette modélisation est de pouvoir ensuite conduire les tâches d’évaluation d’un barrage à un instant donné, de diagnostic des causes de la détérioration de sa sécurité dans le passé et d’analyser la fiabilité et la sécurité du barrage à différentes échelles de temps. Dans cet article, nous proposons une méthode de modélisation basée sur une approche multi-modèles appelée TOM4D (Timed Observations Modeling for Diagnosis) pour le diagnostic du comportement dynamique des barrages. Quatre modèles sont décrits : un modèle de perception, un modèle structurel, un modèle comportemental et un modèle fonctionnel. Nous utilisons ces modèles pour analyser le comportement du barrage à différents instants et en détecter l’éventuelle détérioration. Les modèles résultants de l’étape de modélisation, permet de caractériser et de calculer les diagnostics du barrage. Pour conduire les tâches d’évaluation, diagnostic, deux niveaux sont considérés : les composants individuels (diagnostic local) et le barrage dans sa globalité (diagnostic global). Une application sur un cas réel est effectuée.

[Departement_IRSTEA]Eaux [TR1_IRSTEA]RIVAGE; National audience; L'analyse et le contrôle du comportement des barrages en vue de maîtriser leur fiabilité et leur sécurité nécessitent la mise en ½uvre de modèles prenant en compte le passé (diagnostic), le présent (évaluation) et le futur de l'ouvrage (prédiction et proposition d'actions correctives). Nous avons montré au cours de travaux antérieurs qu'il était possible de modéliser le comportement de l'ouvrage par une approche multi-modèle qui rend possible la prise en compte de ces trois dimensions temporelles. Nous nous proposons dans cet article de montrer l'intérêt de cette modélisation pour l'évaluation des barrages en remblai et le diagnostic de leur comportement. / The analysis and control of dams' behaviour, with the aim of controlling reliability and safety, require the development of models that can take into account the present life of the dam (reliability and safety assessment), its past (diagnosis) and its future (prognosis and corrective actions). Previous works we carried out showed that it is possible to model the dam behaviour using a multi-model approach that takes into account the three temporal dimensions. In this paper, we propose to show the interest of such a modelling for the assessment and diagnosis of embankment dams.

International audience; Dams reliability and safety must be controlled through their whole life to guarantee the safety of the people and assets located at their downstream, to insure they fulfil the functions for which they were built and to safeguard these infrastructures. As a consequence, it is necessary to develop methods and tools for the control of the reliability and safety of dams. In particular, tasks related to the assessment, the diagnosis and the prognosis of dam behaviour are very important to prevent accidents and determine corrective actions. The analysis and control of the dams' behaviour, with the aim of controlling reliability and safety, require the development of models that represent the dynamics of dams. These models have to represent the complexity of the dam behaviour while being easy to handle and to instantiate to the whole set of dams that present different features. They must admit as inputs, miscellaneous data. Finally, these models must be adequate to carry out diagnosis or prognosis tasks to improve dam reliability and safety. In this paper, a multi-model approach is proposed to consider the past, the present and the future behaviour of the dam. Our approach is based on four models: a structural model describing relations between components, a functional model describing the relations which determine the affectation of a possible value to a variable, a behavioural model describing the states of the system and the discrete events that represent the state transitions and a perception model that defines the process and its operating modes. These models were validated by a panel of experts. The method was applied to a real case study concerning a dam which suffered from an internal erosion mechanism. These models were developed to carry out diagnosis, assessment, prognosis and control tasks. Currently, our works show that the developed models are relevant to perform assessment and diagnosis tasks. Moreover, the multi-model approach presents the advantage to facilitate the knowledge representation and handling as well as the results communication to owners or reservoir operators. One of the perspectives is to use the models to forecast dam reliability and safety through time

Forecasting dam behaviour is of paramount importance to avoid collapses. It consists of detecting and controlling deterioration mechanisms. This paper sets out to propose a multi model methodology based on dated observation called TOM4D. The objective is to model dynamic systems, such as dams and to diagnose it. The main idea is to use the same level of abstraction that the Experts in order to reduce the component number to diagnose. This facilitates the diagnosis task. To this aim, the process to be diagnosed is defined with four main models: (1) a perception model that specifies the process as a relation between variables and a set of constraints about the values of process variables, (2) a structural model that describes the relations between the components of the process, (3) a functional model that describes the relations between the values of the process variables (i.e. the functions) and (4) a behavioural model that describes the system states and the discrete events firing the state transitions. These four models contain the fundamental knowledge about the process. They are developed from a knowledge base telling the ageing stories of the process that is to say from scenarios described in terms of events. These four models are used in the TOM4D method aiming at producing some generic models. This method is based on knowledge used by experts and their level of abstraction. This paper shows the obtained generic models about an embankment dam. Using of these models is get onto showing a first overview of using this model to diagnose and forecast the behavior of dam. This explain how using of this method may help the actors in charge of hydraulic engineering to control and diagnose dam., La prévision du comportement des barrages joue un rôle important dans la maîtrise de la sécurité des ouvrages. Elle consiste en la détection et le contrôle des mécanismes de dégradations. Ce papier présente une méthodologie multi modèle basée sur des observations datées, appelé TOM4D. L'objectif est de modéliser des systèmes dynamiques tels que les barrages afin d'en proposer un diagnostic et de prévoir leur comportement. L'idée de base consiste à utiliser le même niveau d'abstraction que celui des experts dans le but de réduire le nombre de composants à diagnostiquer et ainsi de faciliter le diagnostic. Ainsi, le processus à diagnostiquer est défini suivant quatre modèles principaux : (1) un modèle de perception qui spécifie le processus comme une relation entre les variables et un ensemble de contraintes sur les valeurs des variables du processus, (2) un modèle structurel qui décrit les relations entre les composants du processus, (3) un modèle fonctionnel qui décrit les relations entre les valeurs des variables du processus (i.e. des fonctions mathématiques) et (4) un modèle comportemental qui décrit les états du système et les événements discrets initiant les transitions d'état. Ces quatre modèles contiennent la connaissance fondamentale du processus. Ils sont développés à partir d'une base de connaissances contenant l'histoire du vieillissement du processus transformable en scénarios décrits en termes d'événements. Ces quatre modèles sont les principaux modèles de la méthode TOM4D. Cette méthode, visant à produire des modèles génériques du processus, est basée sur la connaissance utilisée par les experts et leur niveau d'abstraction. Ce papier présente les modèles génériques obtenus dans le cas d'un barrage en remblai ainsi qu'un bref aperçu de l'utilisation de ces modèles pour diagnostiquer et prévoir le comportement du barrage. Ceci illustre l'aide qu'apporte l'utilisation de cette méthode aux exploitants, bureaux d'études et experts pour la maîtrise de la sécurité des ouvrages hydrauliques.

International audience; This paper presents a modelling approach of dynamic process for diagnosis that is compatible with the Stochastic Approach framework for discovering temporal knowledge from the timed observations contained in a database. The motivation is to define a multi-model formalism able to represent both the knowledge of these two sources. The aim is to model the process at the same level of abstraction that an expert uses to diagnose the process. The underlying idea is that at this level of abstraction, the model is simple enough to allow an efficient diagnosis. The proposed formalism represents the knowledge in four models: a structural model defining the components and the connection relations of the process, a behavioural model defining the states and the transition states of the process, a functional model containing the logical relations between the values of the process variables, which are defined in the perception model. The models are linked with the process variables. This point facilitates the analysis of the four models consistency and is the basis of the corresponding knowledge modelling methodology. The formalism and methodology are illustrated with the model of a hydraulic dam in Cublize (France).

This paper presents a modeling method of dynamic system like hydraulic dams based on dated observation for diagnosis. This method called TOM4D, proposes to make explicit the implicit models used by experts when they diagnose a system. This method is based on knowledge used by experts and their level of abstraction. This paper shows the obtained generic models about an embankment dam. Using these models may help the actors in charge of hydraulic engineering to control and diagnose dam., Ce papier présente une méthode de modélisation du comportement des barrages appréhendés comme des systèmes dynamiques. Cette méthode, appelée TOM4D, est basée sur des observations datées pour le diagnostic. Elle propose, à partir des connaissances fournies par des experts et en se plaçant à leur niveau d'abstraction, d'expliciter les modèles implicites sur lesquels ces experts raisonnent lors des tâches de diagnostic. Ce papier présente les modèles génériques établis pour les barrages en remblai. L'utilisation de ces modèles devrait aider les exploitants, les bureaux d'études et les experts à mieux contrôler et diagnostiquer le comportement des ouvrages hydrauliques.

This paper proposes the dam behavior modeling with a multi models approach. The expert knowledge wording is pertinent and efficient to a diagnosis task but mixes in a same model knowledge concerning different points of view. So, the multi models approach separates different knowledge natures in three different models. That permits us to monitor, diagnose and predict the behaviour of a dynamic system as a dam. This article shows the three models obtained by the approach application to a homogeneous fill dam which has suffer to internal erosion., Dans cet article, nous proposons de modéliser le comportement des barrages à l'aide d'une approche multi modèles. La formulation des connaissances par les experts, bien qu'adéquats pour un diagnostic efficace et pertinent, mélange souvent en un modèle unique les connaissances relatives à de multiples points de vue. Ainsi, l'approche multi modèles sépare les connaissances expertes de natures différentes en trois modèles qui permettent d'effectuer de la surveillance, du diagnostic et de la prédiction. Cet article présente les trois modèles obtenus par l'application de l'approche à un barrage en remblai homogène ayant subi un mécanisme d'érosion interne.