No description available.

To improve the safety and comfort of intelligent vehicles, advanced driver models offer promising solutions. However, several shortcomings of these models prevent them from being widely applied in reality. To address these shortcomings, advanced artificial intelligence algorithms in conjunction with the sufficient driving environmental factors are proposed based on real-life driving data. More specifically, three typical problems will be addressed in this thesis: Mandatory Lane Changing (MLC) suggestion at the highway entrance; Discretionary Lane Changing (DLC) intention prediction; Car-Following gap model considering the effect of cuts-in from the adjacent lanes. For the MLC suggestion system, in which the main challenges are efficient decision making and high prediction accuracy of both non-merge and merge events, an additional gated branch neural network (GBNN) is proposed. The proposed GBNN algorithm not only achieves the highest accuracy among conventional binary classifiers in terms of great performance on the non-merge accuracy, the merge accuracy, and receiver operating characteristic score but also takes less time. For the DLC, we propose a recurrent neural network (RNN)-based time series classifier with a gated recurrent units (GRU) architecture to predict the surrounding vehicles’ intention. It can predict the surrounding vehicles’ lane changing maneuver 0.8 s in advance at a recall and precision of 99.5% and 98.7%, respectively, which outperforms conventional algorithms such as the Hidden Markov Model (HMM). Finally, drivers are typically faced with two competing challenges when following a preceding vehicle. A method is proposed to address the problem through an overall objective function of car-following gap and velocity. Based on this, seeking the strategic car-following gap translates to finding the optimal solution that minimizes the overall objective function. With the support of field data, the method along with concrete models are instantiated and the application of the method is elaborated. / Thesis / Doctor of Philosophy (PhD) / Lane changing and car following are the two most frequently encountered driving behaviours for intelligent vehicles. Substantial research has been carried out and several prototypes have been developed by universities as well as companies. However, the low accuracy and high computational cost prevent the existing lane changing models from providing safer and more reliable decisions for intelligent vehicles. In the existing car-following models, there are also few models that consider the effects of cut-ins from adjacent lanes which may result in their poor accuracy and efficiency. To address these obstacles, advanced artificial intelligence algorithms combined with sufficient driving environmental factors are proposed due to their promise of providing accurate, efficient, and robust lane changing and car-following models. The main part of this thesis is composed of three journal papers. Paper 1 proposed a gated branch neural network for a mandatory lane changing suggestion system at the on-ramps of highways; paper 2 developed a recurrent neural network time-series algorithm to predict the surrounding vehicles’ discretionary lane changing intention in advance; paper 3 researched the strategic car-following gap model considering the effect of cut-ins from adjacent lanes.

No description available.

Phase behaviour information is essential for the development and optimization of hydrocarbon resource production, transport and refining technologies. Experimental data sets for mixtures containing heavy oil and bitumen are sparse as phase behaviour data are difficult to obtain and cost remains prohibitive for most applications. A computational tool that predicts phase behaviours reliably for mixtures containing such ill-defined components, over broad temperature, pressure and composition ranges would play a central role in the advancement of bitumen production and refining process knowledge and would have favourable impacts on the economics and environmental effects linked to the exploitation of such ill-defined hydrocarbon resources. Prior to this work, predictive computational methods were reliable for dilute mixtures of ill-defined constituents. To include a much wider range of conditions, three major challenges were addressed. The challenges include: creation of a robust and accurate numerical approach, implementation of a reliable thermodynamic model, and speciation of ill-defined constituents like Athabasca Bitumen Vacuum Residue (AVR). The first challenge was addressed by creating a novel computational approach based on a global minimization method for phase equilibrium calculations. The second challenge was tackled by proposing a thermodynamic model that combines the Peng-Robinson equation of state with group contribution and related parameter prediction methods. The speciation challenge was addressed by another research group at the University of Alberta. Pseudo components they proposed were used to assign groups and estimate thermodynamic properties. The new phase equilibrium computational tool was validated by comparing simulated phase diagrams with experimental data for mixtures containing AVR and n-alkanes. There is good qualitative and quantitative agreement between computed and experimental phase diagrams over industrially relevant ranges of compositions, pressures and temperatures. Mismatch was only observed over a limited range of compositions, temperatures and pressures. This computational breakthrough provides, for the first time, a platform for reliable phase behaviour computations with broad potential for application in the hydrocarbon resource sector. The specific computational results can be applied directly to solvent assisted recovery, paraffinic deasphalting, and distillation and refining processes for Athabasca bitumen a strategic resource for Canada. / Chemical Engineering

No description available.

Ce travail porte sur l’analyse et l’étude du comportement de circuits intégrés en technologie CMOS soumis à un impact laser. Une méthodologie d’implémentation d’un impact laser a été développée et améliorée. Ainsi, elle est applicable à n’importe quelle description électrique d’un circuit CMOS, qu’il soit digital ou analogique. Ce procédé est conçu pour permettre aux concepteurs de circuits intégrés pouvant être soumis à des attaques laser, de tester leur circuit en simulation avant leur fabrication et de démontrer leur robustesse.Notre étude s’est focalisée sur le développement d’un outil de simulation intégrant un modèle électrique de l’impact laser sur les transistors MOS afin de reproduire de façon qualitative le comportement du circuit face à un impact laser (attaque semi-invasive en face arrière du circuit), et ce quelques soient ses propriétés physiques.Une première partie d’état de l’art est consacrée à la synthèse des différentes attaques sur circuits sécurisées que l’on peut rencontrer dans le domaine de la microélectronique, telles que les attaques semi-invasives, non invasives ou invasives par exemple. Une seconde partie théorique dédiée à l’interaction laser-silicium au niveau physique nous permet d’étudier les différents acteurs mis en jeu (propriétés physiques du laser – puissance, diamètre et profil du faisceau), avant de les importer comme paramètres dans le domaine électrique.Cette étude se poursuit alors par l’élaboration d’un modèle électrique et d’une méthodologie de simulation dont le but est de permettre de reproduire le comportement de n’importe quel circuit impacté par un laser. Le flot de modélisation passe ainsi en revue l’ensemble des paramètres contrôlables en entrée, qu’il s’agisse des propriétés physiques du laser, traduites dans le domaine électrique, ou encore de la réalité géométrique du circuit impacté, quel que soit sa complexité. Par ailleurs, la flexibilité de cette approche permet de s’adapter à toute évolution du modèle de l’impact laser en lui-même. Il est ainsi possible de simuler un impact intégrant ou non tout ou partie des phénomènes parasites déclenchés par le photocourant. Enfin, il couvre aussi bien des analyses de comportement dans le domaine statique, que dans celui temporel, où la durée d’impulsion du laser prend toute son importance.Afin de démontrer la cohérence de cette méthodologie face à nos attentes théoriques, le comportement de transistors NMOS, PMOS et un inverseur CMOS ont été étudiés au niveau simulation. Cette étude préliminaire nous a permis de calibrer et de valider notre modèle et sa méthodologie d’utilisation avec la théorie attendue: création d’un photocourant proportionnel au potentiel appliqué sur la jonction de drain et couplé au potentiel photoélectrique ainsi qu’à la surface impactée, déclenchement des bipolaires parasites latéraux, etc…. L’analyse sur un inverseur CMOS bufférisé ou non nous donne encore plus d’informations quant aux analyses dynamiques ou statiques : un impact sur un état statique (0 ou 1) ne peut entraîner que des fautes fonctionnelles, alors qu’un impact sur une transition ralentit ou accélère le signal en sortie, au risque de générer une faute fonctionnelle.Enfin, l’étude de différents circuits complexes sur silicium face à plusieurs types de faisceau laser nous a permis de confronter notre méthodologie à la mesure. Une chaîne d’inverseurs, une bascule de type D, et un circuit de verrouillage ont ainsi été impactés. Les résultats observés en simulation sont cohérents avec la mesure, notamment du point de vue comportemental et fonctionnel. / This present work deals with the analysis and study of the integrated circuits behavior in CMOS technology under laser injection. An implementation methodology of a laser impact has been developed and optimized. The study has been focused on the development of a simulation tool integrating an electrical model of a laser impact on MOS transistor. This allows to reproduce in a qualitative way the behavior of a circuit under laser impact (semi-invasive attack on rear face of the circuit), whatever the physical properties of the laser.A preliminary study allowed us to calibrate a new electrical model and its use methodology based on the expected theory: photocurrent creation proportional to the applied potential on the drain junction and linked to the photoelectrical potential with the impacted area; triggering of the lateral parasitic bipolar transistors.The analysis of different complex circuits on silicon under different kind of laser beam allowed us also to validate the developed tool and its implementation methodology: it will help designers to prevent or predict such behavior of their circuits under laser attack, allowing them to find solutions of countermeasures and thus making their integrated circuits more robust in critical applications.