Trois questions importantes se posent à l'élaboration de systèmes d'aide au diagnostic. Quels sont les éléments de la connaissance indispensables au diagnostic? Quelle forme doit prendre la représentation de la connaissance pour être facilement exploitée par l'ingénieur? Comment doit-on organiser ces éléments et quels sont les mécanismes de traitement impliqués qui facilitent l'adaptation du système informatique aux diverses installations rencontrées? C'est à ces questions que cette thèse s'adresse. L'objectif de cette thèse est d'élaborer un mode de représentation de la connaissance qui soit proche des formalismes et des modèles employés par l'ingénieur et qui puisse organiser la connaissance en entités correspondant aux éléments d'une installation à diagnostiquer. Cette représentation de la connaissance repose sur la notion de composante qui regroupe dans une entité les éléments de connaissance relatifs à cette composante. La composante procure la souplesse et rend explicite l'organisation fonctionnelle et structurelle des éléments physiques et conceptuels de l'installation. Chaque composante intègre la description des connaissances relatives aux entrées et sorties, aux paramètres internes, aux comportements, aux modèles de panne, aux fonctions et aux heuristiques de ces éléments. Pour faciliter l'exploitation de la représentation par l'ingénieur, le formalisme exprime les relations algébriques, qualitatives et descriptives des modèles utilisés par celui-ci. Pour ce faire, le formalisme combine les aspects algébriques de la connaissance avec la logique des prédicats, ce qui constitue un des aspects originaux de cette thèse. Ce lien avec la logique des prédicats apporte un support théorique qui met en relation la représentation avec celles présentées par d'autres auteurs du domaine. Cette thèse décrit le formalisme de la représentation et les mécanismes qui résolvent la dimension logique et la dimension algébrique de la connaissance représentée. Les mécanismes parcourent les liens définis entre les éléments de l'installation, tout en conservant les chemins d'inférences employés par ces mécanismes. Un prototype a été élaboré et plusieurs exemples sont résolus par celui-ci. [Résumé abrégé par UMI]

<p>This is a final thesis done at BT, considering one of their forklift truck models called Reflex. The first part of this report is about a preliminary investigation investigating what kind of diagnostic systems BTwants to use, and also which demands there are to meet all expectations on such system. Secondly a diagnostic system, which will show if the drive wheel is worn out, will be presented. </p><p>In the preliminary investigation, two kinds of diagnostic systems were mentioned. These were Model based diagnosis and Predictive analysis. Model based diagnosis is based on measurements made by sensors at the truck, while predictive analysis is based more on statistics and retrieved data about the lifetime of a truck in specific environments. </p><p>The diagnosis system for the drive wheel is based on a model made in Matlab's Simulink. Due to poor documentation, rough simplifications in the model have been made. However, one can still see the differences of principle. </p><p>The main thought was detecting a difference in the lowest torque level from the engine, varying the diameter of the drive wheel. By measurements made directly at the truck, different torques could be observed with varying diameter of the drive wheel, varying load on the truck and varying friction in the gearbox. Using hypothesis tests, it is possible to say whether the drive wheel is worn out or not. </p><p>Results show that if the drive wheel diameter is reduced by 25 mm, torque is reduced by 7% and if the drive wheel diameter is reduced as much as 50 mm, a torque reduction of 11% would be achieved.</p>

<p>Today’s society depends on complex and technically advanced mechanical systems, often containing a variety of different components. Despite careful development andconstruction, some of these components may eventually fail. To avoid unnecessary damage, for example environmental or financial, there is a need to locate and diagnose these faults as fast as possible. This can be done with a diagnostic system, which should produce an alarm if there is a fault in the mechanical system and, if possible, indicate the reason behind it. </p><p>In model based diagnosis, a mathematical model of a fault free system is used to detect if the monitored system contain any faults. This is done by constructing fault indicators, called fault tests, consisting of equations from different parts of the model. Finding these parts is a time-consuming and demanding task, hence it is preferable if as much as possible of this process can be automated. In this thesis an algorithm that finds all parts of a system that can be used to create these fault tests is presented. To make this analysis feasible, in industrial applications, a simplified version of a system model called a structural model is used. Since the models considered in this thesis are implemented in the mathematical software Simulink, a method for transforming Simulink models into analytical equations and structural models is described. As a way of increasing the diagnostic performance for a model based diagnostic system, information about different faults, called fault models, can be included in the model. However, since the models in this thesis are implemented in Simulink, there is no direct way in which this can be preformed. This thesis describes a solution to this problem. The correctness of the algorithms in this thesis are proved and they have been applied, with supreme results, to aScania truck engine model.</p>

This is a final thesis done at BT, considering one of their forklift truck models called Reflex. The first part of this report is about a preliminary investigation investigating what kind of diagnostic systems BTwants to use, and also which demands there are to meet all expectations on such system. Secondly a diagnostic system, which will show if the drive wheel is worn out, will be presented. In the preliminary investigation, two kinds of diagnostic systems were mentioned. These were Model based diagnosis and Predictive analysis. Model based diagnosis is based on measurements made by sensors at the truck, while predictive analysis is based more on statistics and retrieved data about the lifetime of a truck in specific environments. The diagnosis system for the drive wheel is based on a model made in Matlab's Simulink. Due to poor documentation, rough simplifications in the model have been made. However, one can still see the differences of principle. The main thought was detecting a difference in the lowest torque level from the engine, varying the diameter of the drive wheel. By measurements made directly at the truck, different torques could be observed with varying diameter of the drive wheel, varying load on the truck and varying friction in the gearbox. Using hypothesis tests, it is possible to say whether the drive wheel is worn out or not. Results show that if the drive wheel diameter is reduced by 25 mm, torque is reduced by 7% and if the drive wheel diameter is reduced as much as 50 mm, a torque reduction of 11% would be achieved.

Today’s society depends on complex and technically advanced mechanical systems, often containing a variety of different components. Despite careful development andconstruction, some of these components may eventually fail. To avoid unnecessary damage, for example environmental or financial, there is a need to locate and diagnose these faults as fast as possible. This can be done with a diagnostic system, which should produce an alarm if there is a fault in the mechanical system and, if possible, indicate the reason behind it. In model based diagnosis, a mathematical model of a fault free system is used to detect if the monitored system contain any faults. This is done by constructing fault indicators, called fault tests, consisting of equations from different parts of the model. Finding these parts is a time-consuming and demanding task, hence it is preferable if as much as possible of this process can be automated. In this thesis an algorithm that finds all parts of a system that can be used to create these fault tests is presented. To make this analysis feasible, in industrial applications, a simplified version of a system model called a structural model is used. Since the models considered in this thesis are implemented in the mathematical software Simulink, a method for transforming Simulink models into analytical equations and structural models is described. As a way of increasing the diagnostic performance for a model based diagnostic system, information about different faults, called fault models, can be included in the model. However, since the models in this thesis are implemented in Simulink, there is no direct way in which this can be preformed. This thesis describes a solution to this problem. The correctness of the algorithms in this thesis are proved and they have been applied, with supreme results, to aScania truck engine model.

This paper deals with some of the issues that affect the understanding and functioning of descriptive psychopathology (DP). The latter remains the language of description in psychiatry and the basis for her nosological constructs. DP makes assumptions concerning the nature of its object and its underlying causes (i.e. makes use of the medical model). DP needs historical, clinical and numerical recalibration. It is suggested that in some cases, and against traditional psychometric principles, some instruments will have to be constructed that show flexibility and can be varied according to the descriptive needs presented by individual cases. / En el presente artículo se discuten algunos aspectos que afectan a la comprensión y al funcionamiento de la psicopatología descriptiva (PD), la cual proporciona un lenguaje descriptivo a la psiquiatría y las bases para sus constructos nosológicos. La PD formula postulados sobre la naturaleza de su objeto de estudio y sus causas subyacentes (haciendo uso del modelo médico). Se propone que la PD necesita una recalibración histórica, clínica y numérica. En relación a esto, se sugiere que en algunos casos, en contra de los principios psicométricos tradicionales, ciertos instrumentos deberán ser construidos de un modo flexible que permita que varíen de acuerdo a las necesidades descriptivas que presenten casos individuales.

Recent technological advances in the automotive industry have made vehicularsystems increasingly complex in terms of both hardware and software. As thecomplexity of the systems increase, so does the complexity of efficient monitoringof these system. With increasing computational power the field of diagnosticsis becoming evermore focused on software solutions for detecting and classifyinganomalies in the supervised systems. Model-based methods utilize knowledgeabout the physical system to device nominal models of the system to detect deviations,while data-driven methods uses historical data to come to conclusionsabout the present state of the system in question. This study proposes a combinedmodel-based and data-driven diagnostic framework for fault classification,severity estimation and novelty detection. An algorithm is presented which uses a system model to generate a candidate setof residuals for the system. A subset of the residuals are then selected for eachfault using L1-regularized logistic regression. The time series training data fromthe selected residuals is labelled with fault and severity. It is then compressedusing a Gaussian parametric representation, and data from different fault modesare modelled using 1-class support vector machines. The classification of datais performed by utilizing the support vector machine description of the data inthe residual space, and the fault severity is estimated as a convex optimizationproblem of minimizing the Kullback-Leibler divergence (kld) between the newdata and training data of different fault modes and severities. The algorithm is tested with data collected from a commercial Volvo car enginein an engine test cell and the results are presented in this report. Initial testsindicate the potential of the kld for fault severity estimation and that noveltydetection performance is closely tied to the residual selection process.

This dissertation thesis introduces new methods of automated knowledge-base creation and tuning in information and expert systems. The thesis is divided in the two following parts. The first part is focused on the legacy expert system NPS32 developed at the Faculty of Electrical Engineering and Communication, Brno University of Technology. The mathematical base of the system is expression of the rule uncertainty using two values. Thus, it extends information capability of the knowledge-base by values of the absence of the information and conflict in the knowledge-base. The expert system has been supplemented by a learning algorithm. The learning algorithm sets weights of the rules in the knowledge base using differential evolution algorithm. It uses patterns acquired from an expert. The learning algorithm is only one-layer knowledge-bases limited. The thesis shows a formal proof that the mathematical base of the NPS32 expert system can not be used for gradient tuning of the weights in the multilayer knowledge-bases. The second part is focused on multilayer knowledge-base learning algorithm. The knowledge-base is based on a specific model of the rule with uncertainty factors. Uncertainty factors of the rule represents information impact ratio. Using a learning algorithm adjusting weights of every single rule in the knowledge base structure, the modified back propagation algorithm is used. The back propagation algorithm is modified for the given knowledge-base structure and rule model. For the purpose of testing and verifying the learning algorithm for knowledge-base tuning, the expert system RESLA has been developed in C#. With this expert system, the knowledge-base from medicine field, was created. The aim of this knowledge base is verify learning ability for complex knowledge-bases. The knowledge base represents heart malfunction diagnostic base on the acquired ECG (electrocardiogram) parameters. For the purpose of the comparison with already existing knowledge-basis, created by the expert and knowledge engineer, the expert system was compared with professionally designed knowledge-base from the field of agriculture. The knowledge-base represents system for suitable cultivar of winter wheat planting decision support. The presented algorithms speed up knowledge-base creation while keeping all advantages, which arise from using rules. Contrary to the existing solution based on neural network, the presented algorithms for knowledge-base weights tuning are faster and more simple, because it does not need rule extraction from another type of the knowledge representation.