This thesis investigates modeling and simulation of hybrid electric vehicles with particular emphasis on transient modeling and real-time simulation. Three different computer models, i.e. a steady state model, a fully-detailed transient model and a reduced-intensity transient model, are developed for a hybrid drive-train in this study. The steady-state model, which has low computational intensity, is used to determine the optimal battery size and chemistry for a plug-in hybrid drive-train. Simulation results using the developed steady state model show the merits of NiMH and Li-ion battery technologies. Based on the obtained results and the reducing cost of Li-ion batteries, this battery chemistry is used throughout this research. A fully-detailed transient model is developed to simulate the vehicle behaviour under different driving conditions. This model includes the dynamics of the power train components such as the engine, the power-electronic converters and vehicle controllers of all levels. The developed transient model produces an accurate representation of the drive-train including the switching behaviour of the power electronic converters. A reduced-intensity transient model (also referred to as a dynamic average model) is developed for real-time hardware-in-loop simulation of the vehicle. By reducing the computational demand of the detailed transient model using averaging techniques, the reduced-intensity model is implemented on a real-time simulator and is interfaced to an external subsystem such as an actual battery. The setup can be used to test existing and emerging battery technologies, which may not have an accurate mathematical model. Extensive tests are performed to verify the accuracy and validity of the results obtained from the developed hardware-in-loop simulation setup.

This thesis investigates modeling and simulation of hybrid electric vehicles with particular emphasis on transient modeling and real-time simulation. Three different computer models, i.e. a steady state model, a fully-detailed transient model and a reduced-intensity transient model, are developed for a hybrid drive-train in this study. The steady-state model, which has low computational intensity, is used to determine the optimal battery size and chemistry for a plug-in hybrid drive-train. Simulation results using the developed steady state model show the merits of NiMH and Li-ion battery technologies. Based on the obtained results and the reducing cost of Li-ion batteries, this battery chemistry is used throughout this research. A fully-detailed transient model is developed to simulate the vehicle behaviour under different driving conditions. This model includes the dynamics of the power train components such as the engine, the power-electronic converters and vehicle controllers of all levels. The developed transient model produces an accurate representation of the drive-train including the switching behaviour of the power electronic converters. A reduced-intensity transient model (also referred to as a dynamic average model) is developed for real-time hardware-in-loop simulation of the vehicle. By reducing the computational demand of the detailed transient model using averaging techniques, the reduced-intensity model is implemented on a real-time simulator and is interfaced to an external subsystem such as an actual battery. The setup can be used to test existing and emerging battery technologies, which may not have an accurate mathematical model. Extensive tests are performed to verify the accuracy and validity of the results obtained from the developed hardware-in-loop simulation setup.

Produkter som produceras måste genomgå tester och godkännande av kontrollerande organ innan de kan säljas på marknaden. Detta är ofta en tidskrävande och ekonomisk kostsam process att utveckla en ny produkt. Hardware in loop (HIL) är ett sätt att möjliggöra tester av delsystem utan att hela systemet är komplett. Genom att använda HIL under utvecklingen av en ny produkt kan man hitta mjukvarufel samt designfel tidigt i processen, detta kan spara tid och pengar för företag. I bilindustrin började HIL system i form av en körsimulator där föraren kunde få en känsla för bilen utan köra den på vägen. Metoden lämpar sig bra för tester av autonoma drönare då deras funktion behöver valideras med rigorösa tester innan kontrollerande organ tillåter försäljning av produkterna. Risker för personskador eller prototypskador under testning kan minskas då felaktigheter troligtvis hittas och korrigeras tidigare. HIL har varit en del av utvecklingsprocesser de senaste hundra åren och har visat sig varit en effektiv metod. Under detta arbete skapades en simulerad miljö med simuleringsverktyget Carla. Carla är en simulator som bygger på spelmotorn Unreal Engine, den är framtagen med målet att användas för utveckling av tekniker för självkörande bilar. Data från den simulerade miljön användes som insignal till en HIL testbänk. Testbänken kontrollerar insignaler till en kontroll algoritm och övervakar utsignaler från drönaren som kontroll algoritmen styr. Genom att kunna tillhandahålla testbänken med simulerade data är det möjligt att utsätta kontrollalgoritmen för miljöer och omgivningar som vanligtvis hade varit besvärligt och tidskrävande att skapa i verkligheten. Utifrån det kommer vi kunna utsätta drönaren för vår testmiljö och verifiera att den fungerar som designat.

No description available.

Les systèmes piles à combustible de type PEM pour des applications de transport reposent sur un ensemble d’auxiliaires (stockage d’hydrogène, compresseur d’air, convertisseur de puissance, humidificateur, etc) qui assurent le bon fonctionnement du système pile. La mise en place d’observateurs permet de disposer d’un outil pour reconstruire les états non mesurés de ce système; cela permet de mettre en place un contrôle par retour de sortie en vue d’optimiser les performances du système pile et ainsi d'améliorer la détection et l’isolation de défauts (FDI). Cette thèse est basée sur l’étude et la synthèse d'observateurs adaptatifs par mode glissant d’ordre supérieur, pour deux principaux auxiliaires de la pile que sont, le système d'alimentation en air et les convertisseurs de puissance associés à la pile. La première partie de la thèse est consacrée à la synthèse d’observateurs pour la reconstruction des états et à la détection et l’isolation des défauts sur le système d’alimentation en air de la PEMFC. Dans un premier temps, un observateur algébrique par mode glissant d’ordre supérieur est synthétisé pour la reconstruction de la pression partielle de l'oxygène et de l'azote. Dans un deuxième temps, un nouvel observateur adaptatif par mode glissant d’ordre deux est synthétisé pour assurer l'observation simultanée des états, l'identification des paramètres, la surveillance et la reconstruction de défaut dans le circuit d’air. Les performances des observateurs proposées ont été validées grâce à un simulateur Hardware-In-Loop (HIL) du système pile à combustible.Dans la deuxième partie, nous nous sommes intéressés à l’élaboration d’observateurs et de commande par retour de sortie pour les convertisseurs associé au système pile dans une application transport. Ainsi, une commande novatrice par mode glissant d’ordre deux, de type retour de sortie, a été élaborée pour le convertisseur AC/DC. Dans un second temps, un observateur de type modes glissants d’ordre 2 adaptatif est synthétisé pour un convertisseur de type multicellulaire. / Automotive PEM Fuel Cell systems rely upon a set of auxiliary systems for proper operation, such as humidifier, air-feed compressor, power converter etc. The internal physical states of the latter are often unmeasurable, yet required for their precise control. Observers provide a means of obtaining the unmeasured states of these auxiliary systems for feedback control, optimal energy consumption and Fault Diagnosis and Isolation (FDI). This thesis is based on higher order sliding mode observer design studies for two major PEMFC auxiliary systems found in modern automobiles, the air-feed system and the power electronics system.The first part is focused on robust observation and FDI of the PEMFC air-feed systems. Sliding mode observer design and their applications to FDI have been studied in detail for this purpose and the key observation problems in this system have been identified. Based on this study, two solutions are proposed, a sliding mode algebraic observer for oxygen and nitrogen partial pressures and a novel robust adaptive-gain Second Order Sliding Mode (SOSM) observer based FDI for simultaneous state observation, parameter identification, health monitoring and fault reconstruction of the PEMFC air-feed system. The performance of the proposed observers has been validated on an instrumented Hardware-In-Loop (HIL) test bench.The observation and output feedback control problems of different power electronic converters, commonly found in fuel cell vehicles, are addressed in the next part. Robust output feedback SOSM control for three phase AC/DC converters have been presented. A robust SOSM observer for multi-cell converters has also been designed. The performance of all these designs has been demonstrated through a multi-rate simulation approach. The results highlight the robustness of the observers and controllers against parametric uncertainty, measurement noise and external disturbance.

Submitted by Renata Lopes (renatasil82@gmail.com) on 2017-04-25T15:30:19Z No. of bitstreams: 1 rodrigoarrudafelicioferreira.pdf: 7303630 bytes, checksum: 971316c74f1313f54e1c09c837c93e74 (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2017-04-26T12:05:46Z (GMT) No. of bitstreams: 1 rodrigoarrudafelicioferreira.pdf: 7303630 bytes, checksum: 971316c74f1313f54e1c09c837c93e74 (MD5) / Made available in DSpace on 2017-04-26T12:05:46Z (GMT). No. of bitstreams: 1 rodrigoarrudafelicioferreira.pdf: 7303630 bytes, checksum: 971316c74f1313f54e1c09c837c93e74 (MD5) Previous issue date: 2015-03-13 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Este trabalho apresenta um estudo sobre alguns dos aspectos relacionados às microrredes em corrente contínua, uma alternativa para utilização de fontes renováveis de energia em sistemas de geração distribuída. Considerando que a manutenção da tensão no barramento principal das microrredes, no qual as fontes e as cargas são conectadas, é uma das questões mais importantes para a operação de modo satisfatório destes sistemas, um sistema de controle descentralizado de tensão integrando técnicas de controle por droop de tensão e de controle por modos deslizantes é proposto. Uma microrrede CC de 10 kW conectada à rede CA e composta por arranjos fotovoltaicos e bancos de baterias, dimensionada para atender uma parcela das cargas eletrônicas e de iluminação do Instituto Federal de Educação, Ciência e Tecnologia – Campus Juiz de Fora, é modelada e simulada em tempo real utilizando o conceito de simulação CHIL. Resultados experimentais utilizando controladores externos como dispositivos em teste são utilizados para analisar o comportamento do sistema em diferentes condições e para validação da metodologia proposta. Além disso, é apresentado um estudo das técnicas de análise de estabilidade de tensão para sistemas em corrente contínua. Por fim, é apresentada uma metodologia baseada em síntese de elementos reativos utilizando conversores estáticos para estabilização ativa de sistemas CC simplificados, contendo uma fonte e uma carga do tipo potência constante. / This work presents aspects related to DC microgrids, an alternative way of using renewable energy sources in a decentralized fashion. Whereas the maintenance of the microgrid main bus voltage, in which the sources and loads are connected, is one of the most important issues for the satisfactory operation of these systems, a novel methodology for regulating DC bus voltage integrating voltage droop control and sliding mode control techniques is proposed. A grid-connected 10 kW DC microgrid containing photovoltaic arrays and a battery bank, sized to meet a portion of electronic and lighting loads of the Federal Institute of Education, Science and Technology - Campus Juiz de Fora, is modeled and simulated in real time using the concept of CHIL simulation. Experimental results using external controllers as hardware under test are used to analyze system behavior under different conditions and to validate the proposed methodology. Furthermore, a study regarding voltage stability analysis techniques applied to DC systems is presented. Finally, a methodology based on synthesis of reactive elements using static converters for active stabilization of simplified CC systems, containing one source and one constant power load, is presented.

This thesis aims to investigate a fault detection and diagnosis system for automotive cameras using unsupervised learning. 1) Can a front-looking wide-angle camera image dataset be created using Hardware-in-Loop (HIL) simulations? 2) Can an Adversarial Autoencoder (AAE) based unsupervised camera fault detection and diagnosis method be crafted for SPA2 Vehicle Control Unit (VCU) using an image dataset created using Hardware-inLoop? 3) Does using AAE surpass the performance of using Variational Autoencoder (VAE) for the unsupervised automotive camera fault diagnosis model? In the field of camera fault studies, automotive cameras stand out for its complex operational context, particularly in Advanced Driver-Assistance Systems (ADAS) applications. The literature review finds a notable gap in comprehensive image datasets addressing the image artefact spectrum of ADAS-equipped automotive cameras under real-world driving conditions. In this study, normal and fault scenarios for automotive cameras are defined leveraging published and company studies and a fault diagnosis model using unsupervised learning is proposed and examined. The types of image faults defined and included are Lens Flare, Gaussian Noise and Dead Pixels. Along with normal driving images, a balanced fault-injected image dataset is collected using real-time sensor simulation under driving scenario with industrially-recognised HIL setup. An AAE-based unsupervised automotive camera fault diagnosis system using VGG16 as encoder-decoder structure is proposed and experiments on its performance are conducted on both the selfcollected dataset and fault-injected KITTI raw images. For non-processed KITTI dataset, morphological operations are examined and are employed as preprocessing. The performance of the system is discussed in comparison to supervised and unsupervised image partition methods in related works. The research found that the AAE method outperforms popular VAE method, using VGG16 as encoder-decoder structure significantly using 3-layer Convolutional Neural Network (CNN) and ResNet18 and morphological preprocessings significantly ameliorate system performance. The best performing VGG16- AAE model achieves 62.7% accuracy to diagnosis on own dataset, and 86.4% accuracy on double-erosion-processed fault-injected KITTI dataset. In conclusion, this study introduced a novel scheme for collecting automotive sensor data using Hardware-in-Loop, utilised preprocessing techniques that enhance image partitioning and examined the application of unsupervised models for diagnosing faults in automotive cameras. / Denna avhandling syftar till att undersöka ett felupptäcknings- och diagnossystem för bilkameror med hjälp av oövervakad inlärning. De huvudsakliga forskningsfrågorna är om en bilduppsättning från en frontmonterad vidvinkelkamera kan skapas med hjälp av Hardware-in-Loop (HIL)-simulationer, om en Adversarial Autoencoder (AAE)-baserad metod för oövervakad felupptäckt och diagnos för SPA2 Vehicle Control Unit (VCU) kan utformas med en bilduppsättning skapad med Hardware-in-Loop, och om användningen av AAE skulle överträffa prestandan av att använda Variational Autoencoder (VAE) för den oövervakade modellen för felanalys i bilkameror. Befintliga studier om felanalys fokuserar på roterande maskiner, luftbehandlingsenheter och järnvägsfordon. Få studier undersöker definitionen av feltyper i bilkameror och klassificerar normala och felaktiga bilddata från kameror i kommersiella passagerarfordon. I denna studie definieras normala och felaktiga scenarier för bilkameror och en modell för felanalys med oövervakad inlärning föreslås och undersöks. De typer av bildfel som definieras är Lens Flare, Gaussiskt brus och Döda pixlar. Tillsammans med normala bilder samlas en balanserad uppsättning felinjicerade bilder in med hjälp av realtidssensor-simulering under körscenarier med industriellt erkänd HIL-uppsättning. Ett AAE-baserat system för oövervakad felanalys i bilkameror med VGG16 som kodaredekoderstruktur föreslås och experiment på dess prestanda genomförs både på den självinsamlade uppsättningen och felinjicerade KITTI-raw-bilder. För icke-behandlade KITTI-uppsättningar undersöks morfologiska operationer och används som förbehandling. Systemets prestanda diskuteras i jämförelse med övervakade och oövervakade bildpartitioneringsmetoder i relaterade arbeten. Forskningen fann att AAE-metoden överträffar den populära VAEmetoden, genom att använda VGG16 som kodare-dekoderstruktur signifikant med ett 3-lagers konvolutionellt neuralt nätverk (CNN) och ResNet18 och morfologiska förbehandlingar förbättrar systemets prestanda avsevärt. Den bäst presterande VGG16-AAE-modellen uppnår 62,7 % noggrannhet för diagnos på egen uppsättning, och 86,4 % noggrannhet på dubbelerosionsbehandlad felinjicerad KITTI-uppsättning. Sammanfattningsvis introducerade denna studie ett nytt system för insamling av data från bilsensorer med Hardware-in-Loop, utnyttjade förbehandlingstekniker som förbättrar bildpartitionering och undersökte tillämpningen av oövervakade modeller för att diagnostisera fel i bilkameror.

Passenger safety and comfort are important aspects in the process of vehicle development. The world is heading towards developing the safest possible vehicle on the road. Using vehicle motion control functions is one of the ways to enhance vehicle stability. These motion control functions need to be developed in an energy optimised way. By complementing some of the development process with virtual models, both the development time and cost could be minimised. Hence, a sustainable way of control function development could be achieved. In order to verify, validate and calibrate vehicle motion control functions, an accurate model of the virtual vehicle is required. Hence, a research question on how good the virtual model needs to be for the purpose has been addressed. This report suggests a framework in order to determine the capabilities of a virtual vehicle.In this report, a comparison study has been carried out by exciting the real car and virtual model of a Volvo XC90 with a focus of covering the six degrees of freedom (Yaw, pitch, roll, longitudinal, lateral and vertical). A semi automated framework that possesses the capability of automating the testing in a virtual platform has been established. From the test results, the virtual vehicle capabilities were determined. Further, in the second part of the report, an example use case has been considered by taking two calibration sets of Electronic stability control (ESC) system in order to verify the previously established framework.The analysis includes various levels of plant and controller complexity such as Model-in-loop, Software-in-loop and Hardware-in-loop and on two different road surfaces, low friction and high friction. From the observations, the virtual models considered correlates well for the purpose of verification and validation. However, for the purpose of calibration, the models need to be fine-tuned in the virtual platform. Furthermore, the correlation on low friction road surface could be improved by simulating the tests using an advanced tyre model. Overall, this study helps in choosing the correct complexity of various subsystems in a vehicle for the purpose of verification, validation and calibration of vehicle motion control functions. / Passagerarsäkerhet och komfort är viktiga aspekter i utvecklingen av ett fordon. Världen är på väg mot att utveckla säkraste möjliga fordon på vägen. Användning av fordonetse rörelsekontrollfunktioner är ett av sätten att förbättra fordonets stabilitet. Dessa rörelsekontrollfunktioner måste utvecklas på ett energioptimerat sätt. Genom att komplettera en del av utvecklingsprocessen med virtuella modeller kan både utvecklingstid och kostnad minimeras. Därför kan ett hållbart sätt att utveckla funktionerna för kontrollfunktioner uppnås. För att verifiera, validera och kalibrera fordonets rörelsekontrollfunktioner krävs en detaljerad modell av ett virtuellt fordon. Därför har en forskningsfråga om hur bra den virtuella modellen måste vara för ändamålet behandlats. Denna rapport föreslår ett ramverk för att bestämma funktionerna hos virtuella fordon.I denna rapport har en jämförelsestudie genomförts genom att excitera den verkliga bilen och den virtuella modellen av en Volvo XC90 med fokus på att täcka de sex frihetsgraderna (gir, nick, roll, längs, lateral, vertikal). Ett semi-automatiserat ramverk som har förmågan att automatisera testningen i en virtuell plattform har skapats. Från testresultaten bestämdes de virtuella fordonsfunktionerna. Vidare har i den andra delen av rapporten ett exempel på användningsfall beaktats genom att man tar två kalibreringsuppsättningar av ESC-system (Electronic Stability Control) för att verifiera det tidigare etablerade ramverket.Analysen innefattar olika nivåer av modell- och styrenhetskomplexitet såsom Model-in-loop, Software-in-loop och Hardware-in-loop och på två olika vägytor, låg friktion och hög friktion. Enligt observationerna är de virtuella modellerna väl korrelerade för verifiering och validering. För kalibreringen måste dock modellerna finjusteras på den virtuella plattformen. Dessutom kunde korrelationen på lågfriktionsvägytan förbättras genom att simulera testerna med hjälp av en avancerad däckmodell. Sammantaget hjälper den här studien att välja rätt komplexitet hos olika delsystem i ett fordon för verifiering, validering och kalibrering av fordonets rörelsekontrollfunktioner.

The purpose of this thesis is to use automation to create appropriate models in PSS/E with the data from Hardware-in-Loop real-time simulations. With the increase in technology of power electronics, the use of High Voltage Direct Current Technology and Flexible Alternating Current Transmission System devices in the electrical power system have increased tremendously. Static Var Compensators are widely used and it is important to have accurate and reliable models for studies relating to power systems planning and interaction. An automation method is proposed to find the parameters of an SVC model in PSS/E with the data from the Hardware-in- loop real-time simulation of the SVC physical controller using Hypersim. The effect of the SVC on the system under steady state and fault conditions are analyzed with HIL simulation of an SVC physical controller in Hypersim and its corresponding model in PSS/E in the IEEE 14 bus system. The parameters of the SVC model in PSS/E can be effectively varied to bring its response closer to that of the response from HIL simulations in Hypersim. An error function is used as a measure to understand the extent of difference between the model and the physical controller.

abstract: Current hybrid vehicle and/or Fuel Cell Vehicle (FCV) use both FC and an electric system. The sequence of the electric power train with the FC system is intended to achieve both better fuel economies than the conventional vehicles and higher performance. Current hybrids use regenerative braking technology, which converts the vehicles kinetic energy into electric energy instead of wasting it. A hybrid vehicle is much more fuel efficient than conventional Internal Combustion (IC) engine and has less environmental impact The new hybrid vehicle technology with it's advanced with configurations (i.e. Mechanical intricacy, advanced driving modes etc) inflict an intrusion with the existing Thermal Management System (TMS) of the conventional vehicles. This leaves for the opportunity for now thermal management issues which needed to be addressed. Till date, there has not been complete literature on thermal management issued of FC vehicles. The primary focus of this dissertation is on providing better cooling strategy for the advanced power trains. One of the cooling strategies discussed here is the thermo-electric modules. The 3D Thermal modeling of the FC stack utilizes a Finite Differencing heat approach method augmented with empirical boundary conditions is employed to develop 3D thermal model for the integration of thermoelectric modules with Proton Exchange Membrane fuel cell stack. Hardware-in-Loop was designed under pre-defined drive cycle to obtain fuel cell performance parameters along with anode and cathode gas flow-rates and surface temperatures. The FC model, combined experimental and finite differencing nodal net work simulation modeling approach which implemented heat generation across the stack to depict the chemical composition process. The structural and temporal temperature contours obtained from this model are in compliance with the actual recordings obtained from the infrared detector and thermocouples. The Thermography detectors were set-up through dual band thermography to neutralize the emissivity and to give several dynamic ranges to achieve accurate temperature measurements. The thermocouples network was installed to provide a reference signal. The model is harmonized with thermo-electric modules with a modeling strategy, which enables optimize better temporal profile across the stack. This study presents the improvement of a 3D thermal model for proton exchange membrane fuel cell stack along with the interfaced thermo-electric module. The model provided a virtual environment using a model-based design approach to assist the design engineers to manipulate the design correction earlier in the process and eliminate the need for costly and time consuming prototypes. / Dissertation/Thesis / Masters Thesis Technology 2014