Ce travail de thèse concerne le développement d'une méthode de pronostic de défaillance des systèmes de production en série. Une méthode de génération d'un indice de santé brut à partir d'un tenseur de données, appelée Méthode des Points Significatifs a été développée puis validée sur un exemple d'illustration. L'indice généré est ensuite traité par une nouvelle méthode appelée méthode des percentiles, qui permet de générer des profils monotones à partir d'un indice de santé brut. Les profils générés sont ensuite modélisés par un processus Gamma, et la fonction de densité de probabilité agrégée introduite dans ce travail a permis d'estimer le temps de vie résiduel (Remaining Useful Life (RUL)) dans un intervalle de confiance qui assure une marge de sécurité à l'utilisateur industriel. La méthode proposée est appliquée avec succès sur des données expérimentales issues des équipements de production industrielle. / This thesis develops a fault prognosis approach for Discrete Manufacturing Processes. A method of raw health index extraction from a data tensor, called Significant Points was developped and validated on an illustrative example. The generated index is later processed by a new method, called Percentile Method, which allows to generate the monotonic profiles from the raw health index. These profiles are then modelled by a Gamma process, and the aggregate probability density function introduced in this work allowed to estimate the Remaining Useful Life (RUL) in a confidence interval that ensures a safety margin for industrial users. The proposed method is applied successfully on the experimental data of industrial production machines.

Assessment of machine health and prediction of future failures are critical for maintenance decisions. Many of the existing methods use unsupervised techniques to construct health indicators by measuring the disparity between the current state and either the healthy or the faulty states of the system. This approach can work well, but if the resulting health indicators are insufficient there is no easy way to steer the algorithm towards better ones. In this thesis a new method for health indicator construction is investigated that aims to solve this issue. It is based on measuring distance after transforming the sensor data into a new space using a feed-forward neural network. The feed-forward neural network is trained using a multi-objective optimization algorithm, NSGA-II, to optimize criteria that are desired in a health indicator. Thereafter the constructed health indicator is passed into a gated recurrent unit for remaining useful life prediction. The approach is compared to benchmarks on the NASA Turbofan Engine Degradation Simulation dataset and in regard to the size of the neural networks, the model performs relatively well, but does not outperform the results reported by a few of the more recent methods. The method is also investigated on a simulated dataset based on elevator weights with two independent failures. The method is able to construct a single health indicator with a desirable shape for both failures, although the latter estimates of time until failure are overestimated for the more rare failure type. On both datasets the health indicator construction method is compared with a baseline without transformation function and does in both cases outperform it in terms of the resulting remaining useful life prediction error using the gated recurrent unit. Overall, the method is shown to be flexible in generating health indicators with different characteristics and because of its properties it is adaptive to different remaining useful life prediction methods. / Estimering av maskinhälsa och prognos av framtida fel är kritiska steg för underhållsbeslut. Många av de befintliga metoderna använder icke-väglett (unsupervised) lärande för att konstruera hälsoindikatorer som beskriver maskinens tillstånd över tid. Detta sker genom att mäta olikheter mellan det nuvarande tillståndet och antingen de friska eller fallerande tillstånden i systemet. Det här tillvägagångssättet kan fungera väl, men om de resulterande hälsoindikatorerna är otillräckliga så finns det inget enkelt sätt att styra algoritmen mot bättre. I det här examensarbetet undersöks en ny metod för konstruktion av hälsoindikatorer som försöker lösa det här problemet. Den är baserad på avståndsmätning efter att ha transformerat indatat till ett nytt vektorrum genom ett feed-forward neuralt nätverk. Nätverket är tränat genom en multi-objektiv optimeringsalgoritm, NSGA-II, för att optimera kriterier som är önskvärda hos en hälsoindikator. Därefter används den konstruerade hälsoindikatorn som indata till en gated recurrent unit (ett neuralt nätverk som hanterar sekventiell data) för att förutspå återstående livslängd hos systemet i fråga. Metoden jämförs med andra metoder på ett dataset från NASA som simulerar degradering hos turbofan-motorer. Med avseende på storleken på de använda neurala nätverken så är resultatet relativt bra, men överträffar inte resultaten rapporterade från några av de senaste metoderna. Metoden testas även på ett simulerat dataset baserat på elevatorer som fraktar säd med två oberoende fel. Metoden lyckas skapa en hälsoindikator som har en önskvärd form för båda felen. Dock så överskattar den senare modellen, som använde hälsoindikatorn, återstående livslängd vid estimering av det mer ovanliga felet. På båda dataseten jämförs metoden för hälsoindikatorkonstruktion med en basmetod utan transformering, d.v.s. avståndet mäts direkt från grund-datat. I båda fallen överträffar den föreslagna metoden basmetoden i termer av förutsägelsefel av återstående livslängd genom gated recurrent unit- nätverket. På det stora hela så visar sig metoden vara flexibel i skapandet av hälsoindikatorer med olika attribut och p.g.a. metodens egenskaper är den adaptiv för olika typer av metoder som förutspår återstående livslängd.

<p>  </p> <p>The integration of Industry 4.0 technologies such as cyber-physical systems, the internet of things, and artificial intelligence has revolutionized the traditional manufacturing systems, making them smart and digital. Maintenance, a critical component of manufacturing, has been incorporated with data-driven strategies such as prognostic and health management (PHM) to improve production efficiency and reliability. This is achieved by real-time sensing and AI-based modeling, which monitor the health condition of operational equipment for fault detection or failure prediction. The results generated by these models provide crucial support for decision-making processes in manufacturing, ranging from maintenance scheduling to production management.</p> <p>This research focuses on data-driven machine health models based on deep learning in manufacturing systems and explores three directions towards the practical implementation of PHM: model interpretation, model adaptability and robustness enhancement, and cost-benefit analysis of maintenance strategies. In terms of model interpretation, the RNN-LSTM-based model prediction on bearing health estimation was analyzed, and the relationship between the model input and output was investigated. The adoption of the LRP technique improved the explainability of the LSTM model beyond predictive maintenance applications. To enhance model adaptability and robustness, a Transformer-based method was developed for fault diagnosis and novel fault detection, which achieved superior performance compared to conventional fault classification AI-based models. The decision-making aspect of PHM was addressed by conducting a cost-benefit analysis on different maintenance strategies, which provided a new perspective for decision-makers in maintenance management.</p>

The growing demand for sustainable technology has led to an increased application of power electronics. As these devices are often exposed to harsh conditions, their reliability is a primary concern for both manufacturers and users. Addressing these reliability challenges involves a set of activities known as Prognostics and Health Management (PHM). In PHM, predicting the Remaining Useful Life (RUL) is crucial. This prediction relies on identifying failure precursors, which signify the presence of degradation. These precursors are then used to construct a degradation model that enables the prediction of the remaining time that the device can work before failure. The project focuses on examining a MOSFET aging dataset from the NASA PCoE dataset depository and a diode aging dataset from Fraunhofer ENAS. The prediction of the remaining useful life of devices using failure precursors has been done by applying recurrent neural network (RNN) methods. However, the prediction results from a single feature is significantly deviated from the actual values. To improve the prediction, the age of the device was proposed as an additional feature. RNNs with a similar number of weights and RNNs with the same hyperparameters are implemented and their performance is evaluated by the accuracy of prediction. The results show that all the RNN models implemented manage to capture the characteristics of the aging data. Despite its simpler structure, the vanilla RNN manages to produce a comparable result with the GRU and LSTM by simpler mechanism and less number of weights. The results also reveal that the characteristics of the data have a significant impact on the final results. / Den växande efterfrågan på hållbar teknik har lett till en ökad tillämpning av kraftelektronik. Eftersom dessa enheter ofta utsätts för tuffa förhållanden är deras tillförlitlighet ett primärt bekymmer för både tillverkare och användare. Att ta itu med dessa tillförlitlighetsutmaningar innebär en uppsättning aktiviteter som kallas Prognostics and Health Management (PHM). I PHM är det avgörande att förutsäga det återstående användbara livet (RUL). Denna förutsägelse bygger på identifiering av felprekursorer, som anger förekomsten av nedbrytning. Dessa prekursorer används sedan för att konstruera en nedbrytningsmodell som möjliggör förutsägelse av den återstående tiden som enheten kan fungera innan fel. Projektet fokuserar på att undersöka en MOSFET-åldringsdataset från NASA PCoE-datauppsättningen och en diodåldringsdataset från Fraunhofer ENAS. Förutsägelsen av den återstående livslängden för enheter som använder felprekursorer har gjorts genom att använda metoder för återkommande neurala nätverk (RNN). Förutsägelseresultatet från en enskild funktion avviker dock avsevärt från de faktiska värdena. För att förbättra förutsägelsen föreslogs enhetens ålder som en extra funktion. RNN med ett liknande antal vikter och RNN med samma hyperparametrar implementeras och deras prestanda utvärderas av förutsägelsens noggrannhet. Resultaten visar att alla implementerade RNN-modeller lyckas fånga egenskaperna hos åldrande data. Trots sin enklare struktur lyckas vanilj RNN producera ett jämförbart resultat med GRU och LSTM genom enklare mekanism och färre antal vikter. Resultaten visar också att uppgifternas egenskaper har en betydande inverkan på de slutliga resultaten.

In recent years, extensive research efforts have been dedicated to the field of prognostics and age-related degradation, with major focus on higher complexity devices. However, relatively little attention has been given to power devices, such as Insulated Bipolar Gate Transistors (IGBTs), despite their critical role in high power electronic applications. These device find their application in various domains, including power grids, where their capability of operating over a broad spectrum of current and voltage levels is a necessity. Because of their central role, their condition can heavily effect the entire system, and the lack of comprehensive understanding and accurate aging prediction for IGBTs poses a significant challenge in ensuring their optimal performance, the deployment of intelligent equipment maintenance and in minimizing the risk of failure. To overcome this research and knowledge gap, the present study focuses on the development and implementation of a Convolutional-Long Short-Term Memory Neural Network, for predicting the value of the component temperature, as the main precursor for its premature aging. Moreover, an incremental learning approach is employed to address the challenges of online learning in real-world scenarios. To evaluate the proposed methodology, a comparative analysis is conducted against a base Long Short-Term Memory (LSTM) model, using an IGBT data set from the NASA Ames Laboratory. The empirical experiments yield promising results, demonstrating that the proposed model outperforms the base LSTM model in terms of accuracy and predictive capabilities. Moreover, the incremental approach appears to be suitable to extend the Convolutional-LSTM model to online learning settings. The findings of this research provide valuable insight into prognostics of power devices and contribute to broaden the field of predictive maintenance, especially in the context of power devices. / Under de senaste åren har omfattande forskningsinsatser ägnats åt prognostik och åldersrelaterad degradering, med fokus på mer komplexa enheter. Kraftelektronik, t.ex. IGBT-transistorer (Insulated Bipolar Gate Transistors), har dock ägnats relativt lite uppmärksamhet, trots deras kritiska roll i elektroniska applikationer med hög effekt. Dessa enheter används inom olika områden, bland annat kraftnät, där deras förmåga att arbeta över ett brett spektrum av ström- och spänningsnivåer är en nödvändighet. På grund av deras centrala roll kan deras tillstånd kraftigt påverka hela systemet, och bristen på omfattande förståelse och exakta åldringsprognoser för IGBT utgör en betydande utmaning för att säkerställa optimal prestanda, implementering av intelligent underhåll av utrustning och för att minimera risken för fel. För att överbrygga denna forsknings- och kunskapslucka fokuserar den här studien på utveckling och implementering av ett neuralt nätverk med faltning och långt korttidsminne för att förutsäga värdet på komponenttemperaturen, som den viktigaste föregångaren till dess för tidiga åldrande. Dessutom används en inkrementell inlärningsmetod för att hantera utmaningarna med online-inlärning i verkliga scenarier. För att utvärdera den föreslagna metoden genomförs en jämförande analys mot en basmodell för Long Short-Term Memory (LSTM), med hjälp av en IGBT-datauppsättning från NASA Ames Laboratory. De empiriska experimenten ger lovande resultat och visar att den föreslagna modellen överträffar den grundläggande LSTM-modellen när det gäller noggrannhet och prediktiva förmågor. Dessutom verkar det inkrementella tillvägagångssättet vara lämpligt för att utvidga Convolutional-LSTM-modellen till onlineinlärningsinställningar. Resultaten av denna forskning ger värdefull insikt i prognostik av kraftaggregat och bidrar till att bredda området för prediktivt underhåll, särskilt i samband med kraftaggregat.

<p>Terminal airspace around an airport is the biggest bottleneck for commercial operations in the National Airspace System (NAS). In order to prognosticate the safety status of the terminal airspace, effective prediction of the airspace evolution is necessary. While there are fixed procedural structures for managing operations at an airport, the confluence of a large number of aircraft and the complex interactions between the pilots and air traffic controllers make it challenging to predict its evolution. Modeling the high-dimensional spatio-temporal interactions in the airspace given different environmental and infrastructural constraints is necessary for effective predictions of future aircraft trajectories that characterize the airspace state at any given moment. A novel deep learning architecture using Graph Neural Networks is proposed to predict trajectories of aircraft 10 minutes into the future and estimate prog?nostic metrics for the airspace. The uncertainty in the future is quantified by predicting distributions of future trajectories instead of point estimates. The framework’s viability for trajectory prediction and prognosis is demonstrated with terminal airspace data from Dallas Fort Worth International Airport (DFW). </p>

A problem in self-driving vehicle (SDV) development is replacing human intuition in the diagnostic process. Some fundamental interactions between driver, service personnel, and system developer are hard to replace by onboard systems and processes. One solution to this problem is to have a staffed control tower that supports the vehicle’s decision-making. In this thesis, a decision-making process for short-term fault avoidance and uptime maximization was developed. A system architecture was proposed and implemented on the SVEA platform. By integrating the onboard system with a control tower, an increase in safe operation was achieved when the vehicle lacked knowledge. In addition, some critical interactions between SDV and control tower were tested: Diagnosis verification and plan correction. By communicating onboard data such as system warnings, symptoms, speed, and location, the vehicle could support the control tower in its decision-making. One conclusion from the thesis was that the SDV with a control tower lowered the threshold for vehicle autonomy. Also, it was shown that both vehicle safety and uptime could be considered in the route planning of SDV:s. In the future, the diagnostic and prognostic algorithms employed in the proposed architecture could be integrated with machine learning tools to update degradation models online. This could make their outputs more reliable and accurate and ultimately make the whole system more safe and reliable. / Ett problem i utvecklingen av självkörande fordon är hur man bäst ersätter den mänskliga intuitionen i diagnosprocessen. Många av nyckelinteraktionerna mellan förare, verkstadspersonal och utvecklingsingenjörer är svåra att ersätta med autonoma processer. En lösning på detta problem är att ha ett kontrolltorn som ger stöd till fordonets beslutsfattande. I det här examensarbetet föreslås en beslutsfattandeprocess för felhantering och uptime-maximering på kort sikt, under körning. En systemarkitektur utvecklades och implementerades på SVEA-plattformen. Genom att integrera systemen i fordonet med ett kontrolltorn kunde en säkrare körning säkerställas i situationer där fordonet saknade relevant kunskap. Några nyckelinteraktioner testades även: Diagnosverifikation och beslutskorrigering. Genom att kommunicera relevant data till kontrolltornet så som systemvarningar, symptom, hastighet och position kunde fordonet även stödja människan i dess beslutandeprocess. En slutsats från arbetet var att detta föreslagna system, självkörande fordon med kontrolltorn, sänkte tröskeln för autonomi i fordon. Det visades också hur både fordonets säkerhet och uptime kan användas som parametrar i ruttplanering för självkörande fordon. I framtiden skulle de framtagna diagnos och prognosalgoritmerna kunna integreras med maskininlärningsverktyg för att möjliggöra live uppdatering av bl.a. degraderingsmodeller. Detta skulle göra dem mer tillförlitliga och precisa vilket i slutändan gör systemet som helhet mer säkert och tillförlitligt.

Remaining useful life (RUL) prediction is a problem that researchers in the prognostics and health management (PHM) community have been studying for decades. Both physics-based and data-driven methods have been investigated, and in recent years, deep learning has gained significant attention. When sufficiently large and diverse datasets are available, deep neural networks can achieve state-of-the-art performance in RUL prediction for a variety of systems. However, for end users to trust the results of these models, especially as they are integrated into safety-critical systems, RUL prediction uncertainty must be captured. This work explores an approach for estimating both epistemic and heteroscedastic aleatoric uncertainties that emerge in RUL prediction deep neural networks and demonstrates that quantifying the overall impact of these uncertainties on predictions reveal valuable insight into model performance. Additionally, a study is carried out to observe the effects of RUL truth data augmentation on perceived uncertainties in the model.

La maintenance prédictive joue un rôle important dans le maintien des systèmes de production continue car elle peut aider à réduire les interventions inutiles ainsi qu'à éviter des pannes imprévues. En effet, par rapport à la maintenance conditionnelle, la maintenance prédictive met en œuvre une étape supplémentaire, appelée le pronostic. Les opérations de maintenance sont planifiées sur la base de la prédiction des états de détérioration futurs et sur l'estimation de la vie résiduelle du système. Dans le cadre du projet européen FP7 SUPREME (Sustainable PREdictive Maintenance for manufacturing Equipment en Anglais), cette thèse se concentre sur le développement des modèles de détérioration stochastiques et sur des méthodes d'estimation de la vie résiduelle (Remaining Useful Life – RUL en anglais) associées pour les adapter aux cas d'application du projet. Plus précisément, les travaux présentés dans ce manuscrit sont divisés en deux parties principales. La première donne une étude détaillée des modèles de détérioration et des méthodes d'estimation de la RUL existant dans la littérature. En analysant leurs avantages et leurs inconvénients, une adaptation d’une approche de l'état de l'art est mise en œuvre sur des cas d'études issus du projet SUPREME et avec les données acquises à partir d’un banc d'essai développé pour le projet. Certains aspects pratiques de l’implémentation, à savoir la question de l'échange d'informations entre les partenaires du projet, sont également détaillées dans cette première partie. La deuxième partie est consacrée au développement de nouveaux modèles de détérioration et les méthodes d'estimation de la RUL qui permettent d'apporter des éléments de solutions aux problèmes de modélisation de détérioration et de prédiction de RUL soulevés dans le projet SUPREME. Plus précisément, pour surmonter le problème de la coexistence de plusieurs modes de détérioration, le concept des modèles « multi-branche » est proposé. Dans le cadre de cette thèse, deux catégories des modèles de type multi-branche sont présentées correspondant aux deux grands types de modélisation de l'état de santé des système, discret ou continu. Dans le cas discret, en se basant sur des modèles markoviens, deux modèles nommés Mb-HMM and Mb-HsMM (Multi-branch Hidden (semi-)Markov Model en anglais) sont présentés. Alors que dans le cas des états continus, les systèmes linéaires à sauts markoviens (JMLS) sont mis en œuvre. Pour chaque modèle, un cadre à deux phases est implémenté pour accomplir à la fois les tâches de diagnostic et de pronostic. A travers des simulations numériques, nous montrons que les modèles de type multi-branche peuvent donner des meilleures performances pour l'estimation de la RUL par rapport à celles obtenues par des modèles standards mais « mono-branche ». / Predictive maintenance plays a crucial role in maintaining continuous production systems since it can help to reduce unnecessary intervention actions and avoid unplanned breakdowns. Indeed, compared to the widely used condition-based maintenance (CBM), the predictive maintenance implements an additional prognostics stage. The maintenance actions are then planned based on the prediction of future deterioration states and residual life of the system. In the framework of the European FP7 project SUPREME (Sustainable PREdictive Maintenance for manufacturing Equipment), this thesis concentrates on the development of stochastic deterioration models and the associated remaining useful life (RUL) estimation methods in order to be adapted in the project application cases. Specifically, the thesis research work is divided in two main parts. The first one gives a comprehensive review of the deterioration models and RUL estimation methods existing in the literature. By analyzing their advantages and disadvantages, an adaption of the state of the art approaches is then implemented for the problem considered in the SUPREME project and for the data acquired from a project's test bench. Some practical implementation aspects, such as the issue of delivering the proper RUL information to the maintenance decision module are also detailed in this part. The second part is dedicated to the development of innovative contributions beyond the state-of-the-are in order to develop enhanced deterioration models and RUL estimation methods to solve original prognostics issues raised in the SUPREME project. Specifically, to overcome the co-existence problem of several deterioration modes, the concept of the "multi-branch" models is introduced. It refers to the deterioration models consisting of different branches in which each one represent a deterioration mode. In the framework of this thesis, two multi-branch model types are presented corresponding to the discrete and continuous cases of the systems' health state. In the discrete case, the so-called Multi-branch Hidden Markov Model (Mb-HMM) and the Multi-branch Hidden semi-Markov model (Mb-HsMM) are constructed based on the Markov and semi-Markov models. Concerning the continuous health state case, the Jump Markov Linear System (JMLS) is implemented. For each model, a two-phase framework is carried out for both the diagnostics and prognostics purposes. Through numerical simulations and a case study, we show that the multi-branch models can help to take into account the co-existence problem of multiple deterioration modes, and hence give better performances in RUL estimation compared to the ones obtained by standard "single branch" models.

La maintenance aéronautique est fortement régulée, notamment à travers l’établissement d’un planning de maintenance obligatoire, permettant de garantir la sureté structurale. La fréquence des arrêts en maintenance est déterminée de manière très conservative en vue d’assurer les exigences de fiabilité. Développer des stratégies de maintenance moins conservatives et plus efficaces peut alors représenter une voie pour une nouvelle croissance des compagnies aériennes. Les systèmes de monitoring embarqué de structures, sont progressivement introduits dans l’industrie aéronautique. Ces développements pourraient alors permettre de nouvelles stratégies de maintenance structurale basées sur la prévision de l’état de santé de chaque élément structural, plutôt que basée sur une maintenance programmée, tel qu’implémentée actuellement. Dans ce cadre général, ce travail se concentre sur le suivi par un système embarqué de la propagation de fissures de fatigue dans les panneaux de fuselage. Une nouvelle méthode de prévision des fissures basée sur des modèles de propagation est développée, qui permet de filtrer le bruit des mesures du système embarqué, identifier la taille actuelle de la fissure et prédire son évolution future et par conséquent la fiabilité des panneaux. Cette approche prédictive est intégrée dans le processus de maintenance structurale aéronautique et deux types de maintenances prédictives sont proposés. L’étude numérique montre que ces stratégies de maintenance prédictive peuvent réduire de manière significative les coûts de maintenance en réduisant le nombre d’arrêts en maintenance et le nombre de réparations inutiles. / Aircraft maintenance represents a major economic cost for the aviation industry. Traditionally, the aircraft maintenance is highly regulated based on fixed schedules (thus called scheduled maintenance) in order to ensure safety. The frequency of scheduled maintenance is designed to be very conservative to maintain a desirable level of reliability. Developing efficient maintenance can be an important way for airlines to allow a new profit growth. With the development of sensor technology, structural health monitoring (SHM) system, which employ a sensor network sealing inside aircraft structures to monitor the damage state, are gradually being introduced in the aviation industry. Once it is possible to monitor the structure damage state automatically and continuously by SHM systems, it enables to plan the maintenance activities according to the actual or predicted health state of the aircraft rather than a fixed schedule. This work focus on the fatigue crack propagation in the fuselage panels. The SHM system is assumed to be employed. A model-based prognostics method is developed, which enables to filter the noise of SHM data to estimate the crack size, and to predict the future health state of the panels. This predictive information is integrated into the maintenance decision-making and two types of predictive maintenance are developed. The numerical study shows that the predictive maintenance significantly reduces the maintenance cost by reducing the number of maintenance stop and the repaired panels.