<p dir="ltr">Progress in deep learning methodologies has transformed the landscape of medical image analysis, opening fresh pathways for precise and effective diagnostics. Currently, multi-phase liver CT scans follow a four-stage process, commencing with an initial scan carried out before the administration of <a href="" target="_blank">intravenous (IV) contrast-enhancing material</a>. Subsequently, three additional scans are performed following the contrast injection. The primary objective of this research is to automate the analysis and prediction of 50% of liver CT scans. It concentrates on discerning patterns of intensity change during the second, third, and fourth phases concerning the initial phase. The thesis comprises two key sections. The first section employs the non-contrast phase (first scan), late hepatic arterial phase (second scan), and portal venous phase (third scan) to predict the delayed phase (fourth scan). In the second section, the non-contrast phase and late hepatic arterial phase are utilized to predict both the portal venous and delayed phases. The study evaluates the performance of two deep learning models, U-Net and U²-Net. The process involves preprocessing steps like subtraction and normalization to compute contrast difference images, followed by post-processing techniques to generate the predicted 2D CT scans. Post-processing steps have similar techniques as in preprocessing but are performed in reverse order. Four fundamental evaluation metrics, including <a href="" target="_blank">Mean Absolute Error (MAE), Signal-to-Reconstruction Error Ratio (SRE), Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure (SSIM), </a>are employed for assessment. Based on these evaluation metrics, U²-Net performed better than U-Net for the prediction of both portal venous (third) and delayed (fourth) phases. Specifically, U²-Net exhibited superior MAE and PSNR results for the predicted third and fourth scans. However, U-Net did show slightly better SRE and SSIM performance in the predicted scans. On the other hand, for the exclusive prediction of the fourth scan, U-Net outperforms U²-Net in all four evaluation metrics. This implementation shows promising results which will eliminate the need for additional CT scans and reduce patients’ exposure to harmful radiation. Predicting 50% of liver CT volumes will reduce exposure to harmful radiation by half. The proposed method is not limited to liver CT scans and can be applied to various other multi-phase medical imaging techniques, including multi-phase CT angiography, multi-phase renal CT, contrast-enhanced breast MRI, and more.</p>

Design of atomizers for gas-turbine purposes are an important ordeal. The per-formance of the atomizer directly impacts the efficiency of the gas-turbine, andconsequently, the energy extracted from the turbine. Furthermore, the design ofthe atomizer can have an impact on reducing toxic emissions. On a global scale,gas-turbines can be considered crucial for the transition to renewable energy. Forengineers, designing of atomizers are however challenging. Turbulent flow, multi-phase interaction and chemical reactions are some of the complex physics involvedwhich has to be taken into consideration when designing the atomizer. Engineerstraditionally uses experimental testing for investigation of designs, and it is still verymuch a useful methodology. However, numerical simulations and CFD have recentlygained popularity due to being a more cost-effective methodology. In this work, theprocedure for designing a prefilm atomizer involving CFD (single phase model andmulti phase model) and experimental testing is documented. The details of the twonumerical models (a single phase model and a multi phase model) has been doc-umented as well as the experimental setup. The single phase model is used for aparametric study and experimental testing is used for evaluation of designs. Themulti phase model is aimed to replicate the experimental results. The validity ofthe numerical models and the experimental setup are discussed, and possible mod-ifications of the methodology for future studies are suggested. Finally, suggestionsfor how the prefilmer should be designed is suggested.

No description available.

No description available.

This dissertation studies modeling and control issues of parallel three-phase pulse-width modulated (PWM) converters. The converters include three-phase boost rectifiers, voltage source inverters, buck rectifiers and current source inverters. The averaging of the parallel converters is performed based on a generic functional switching unit, which is called a phase leg in boost rectifiers and voltage source inverters, and a rail arm in buck rectifiers and current source inverters. Based on phase-leg and rail-arm averaging, the developed models are not only equivalent to the conventional three-phase converter models that are based on phase-to-phase averaging, but they also preserve common-mode information, which is critical in the analysis of the parallel converters. The models reveal such parallel dynamics as reactive power circulation and small-signal interaction. A unique feature of the parallel three-phase converters is a zero-sequence circulating current. This work proposes a novel zero-sequence control concept that uses variable zero-vectors in the space-vector modulation (SVM) of the converters. The control can be implemented within an individual converter and is independent from the other control loops for the converter. Therefore, it greatly facilitates the design and expansion of a parallel system. Proper operation of the parallel converters requires an explicit load-sharing mechanism. In order to have a modular design, a droop method is recommended. Traditionally, however, a droop method has to compromise between voltage regulation and load sharing. After parametric analysis, a novel droop method using a gain-scheduling technique is proposed. The numeric analysis shows that the proposed droop method can achieve both good voltage regulation and good load sharing. An interleaving technique is often used in parallel converter systems in order to reduce current ripples. Because of its symmetrical circuit structure, the parallel three-phase converter system can reduce both differential-mode and common-mode noise with a center-aligned symmetrical SVM. Based on the concept that a symmetrical circuit can reduce common-mode dv/dt noise, a conventional three-phase, four-leg inverter is modified so that its fourth leg is symmetrical to the other three legs. The common-mode dv/dt noise can be practically eliminated with a new modulation strategy. Meanwhile, with a modified control design, the new four-leg inverter still can handle low-frequency common-mode components that occur due to unbalanced and nonlinear load. / Ph. D.

The introduction of communication technology into the electric power grid has made the grid more reliable. Power system operators gain visibility over the power system and are able to resolve operational issues remotely via Supervisory Control And Data Acquisition (SCADA) technology. This reduces outage periods. Nonetheless, the remote-control capability has rendered the power grid vulnerable to cyberattacks. In December 2015, over 200,000 people in Ukraine became victims of the first publicly reported cyberattack on the power grid. Consequently, cyber-physical security research for the power system as a critical infrastructure is in critical need. Research on cybersecurity for power grids has produced a diverse literature; the multi-faceted nature of the grid makes it vulnerable to different types of cyberattacks, such as direct power grid, supply chain and ransom attacks. The attacks may also target different levels of grid operation, such as the transmission system, distribution system, microgrids, and generation. As these levels are characterized by varying operational constraints, the literature may be categorized not only according to the type of attack it targets, but also according to the level of power system operation under consideration. It is noteworthy that cybersecurity research for the transmission system dominates the literature, although the distribution system is noted to have a larger attack surface. For the distribution system, a notable attack type is the so-called direct switching attack, in which an attacker aims to disrupt power supply by compromising switching devices that connect equipment such as generators, and power grid lines. To maximize the damage, this attack tends to be coordinated as the attacker optimally selects the nodes and switches to attack. This decision-making process is often a bi- or tri-level optimization problem which models the interaction between the attacker and the power system defender. It is necessary to detect attacks and establish coordination/correlation among them. Determining coordination is a necessary step to predict the targets of an attack before attack completion, and aids in the mitigation strategy that ensues. While the literature has addressed the direct switching attack on the distribution system in different ways, there are also shortcomings. These include: (i) techniques to establish coordination among attacks are centralized, making them prone to single-point failures; (ii) techniques to establish coordination among attacks leverage only power system models, ignoring the influence of communication network vulnerabilities and load criticality in the decisions of the attacker; (iii) attacker-defender optimization models assume specific knowledge of the attacker resources and constraints by the defender, a strong unrealistic assumption that reduces their usability; (iv) and, mitigation strategies tend to be static and one-sided, being implemented only at the physical level, or at the communication network level. In light of this, this dissertation culminates in major contributions concerning real-time decentralized correlation of detected direct switching attacks and hybrid mitigation for electric power distribution systems. Concerning this, four novel contributions are presented: (i) a framework for decentralized correlation of attacks and mitigation; (ii) an attacker-defender optimization model that accounts for power system laws, load criticality, and cyber vulnerabilities in the decision-making process of the attacker; (iii) a real-time learning-based mechanism for determining correlation among detected attacks and predicting attack targets, and which does not assume knowledge of the attacker's resources and constraints by the power system defender; (iv) a hybrid mitigation strategy optimized in real-time based on information learned from detected attacks, and which combines both physical level and communication network level mitigation. Since the execution of intrusion detection systems and mechanisms such as the ones proposed in this dissertation may deter attackers from directly attacking the power grid, attackers may perform a supply chain cyberattack to yield the same results. Although, supply chain cyberattacks have been acknowledged as potentially far-reaching, and compliance directives put forward for this, the detection of supply chain cyberattacks is in a nascent stage. Consequently, this dissertation also proposes a novel method for detecting supply chain cyberattacks. To the best of the knowledge of the author, this work is the first preliminary work on supply chain cyberattack detection. / Doctor of Philosophy / The electric power grid is the network that transports electricity from generation to consumers, such as homes and factories. The power grid today is highly remote-monitored and controlled. Should there be a fault on the grid, the human operator, often remotely located, may only need to resolve it by sending a control signal to telemetry points, called nodes, via a communication network. This significantly reduces outage periods and improves the reliability of the grid. Nonetheless, the high level connectivity also exposes the grid to cyberattacks. The cyber connectivity between the power grid and the human operator, like all communication networks, is vulnerable to cyberattacks that may allow attackers to gain control of the power grid. If and when successful, wide-spread and extended outages, equipment damage, etc. may ensue. Indeed, in December 2015, over 200,000 people in Ukraine became victims to the first publicly reported cyberattack on a power grid. As a critical infrastructure, cybersecurity for the power grid is, therefore, in critical need. Research on cybersecurity for power grids has produced a diverse literature; the multi-faceted nature of the grid makes it vulnerable to different types of cyberattacks, such as direct power grid, supply chain and ransom attacks. Notable is the so-called direct switching attack, in which an attacker aims to compromise the power grid communication network in order to toggle switches that connect equipment such as generators, and power grid lines. The aim is to disrupt electricity service. To maximize the damage, this attack tends to be coordinated; the attacker optimally selects several grid elements to attack. Thus, it is necessary to both detect attacks and establish coordination among them. Determining coordination is a necessary step to predict the targets of an attack before attack completion. This aids the power grid owner to intercept and mitigate attacks. While the literature has addressed the direct switching attack in different ways, there are also shortcomings. Three outstanding ones are: (i) techniques to determine coordination among attacks and predict attack targets are centralized, making them prone to single-point failures; (ii) techniques to establish coordination among attacks leverage only power system physical laws, ignoring the influence of communication network vulnerabilities in the decisions of the attacker; (iii) and, studies on the interaction between the attacker and the defender (i.e., power grid owner) assume specific knowledge of the attacker resources and constraints by the defender, a strong unrealistic assumption that reduces their usability. This research project addresses several of the shortcomings in the literature, particularly the aforementioned. The work focuses on the electric distribution system, which is the power grid that connects directly to consumers. Indeed, this choice is ideal, as the distribution system has a larger attack surface than other parts of the grid and is characterized by computing devices with more constrained computational capability. Thus, adaptability to simple computing devices is a priority. The contributions of this dissertation provide leverage to the power grid owner to intercept and mitigate attacks in a resilient manner. The original contributions of the work are: (i) a novel realistic model that shows the decision making process of the attacker and their interactions with the defender; (ii) a novel decentralized mechanism for predicting the targets of coordinated cyberattacks on the electric distribution grid in real-time and which is guided by the attack model, (iii) and a novel hybrid optimized mitigation strategy that provides security to the power grid at both the communication network level and the physical power grid level. Since the power grid is constructed with smart equipment from various vendors, attackers may launch effective attacks by compromising the devices deployed in the power grid through a compromised supply chain. By nature, such an attack is evasive to traditional intrusion detection systems and algorithms such as the aforementioned. Therefore, this work also provides a new method to defend the grid against supply chain attacks, resulting in a mechanism for its detection in a critical power system communication device.

Extremely small pore size, low porosity, and ultra-low permeability are among the characteristics of shale rocks. In tight shale reservoirs, the nano-confinement effects that include large gas-oil capillary pressure and critical property shifts could alter the phase behaviors, thereby affecting the oil or gas production. In this research, two in-house simulation models, i.e., a compositionally extended black-oil model and a fully composition model are developed to examine the nano-pore confinement effects on convective and diffusive multicomponent multiphase fluid transport. Meanwhile, the effect of nano-confinement and rock intrinsic properties (porosity and tortuosity factor) on predicting effective diffusion coefficient are investigated. First, a previously developed compositionally extended black-oil simulation approach is modified, and extended, to include the effect of large gas-oil capillary pressure for modeling first contact miscible (FCM), and immiscible gas injection. The simulation methodology is applied to gas flooding in both high and very low permeability reservoirs. For a high permeability conventional reservoir, simulations use a five-spot pattern with different reservoir pressures to mimic both FCM and immiscible displacements. For a tight oil-rich reservoir, primary depletion and huff-n-puff gas injection are simulated including the effect of large gas-oil capillary pressure in flow and in flash calculation on recovery estimations. A dynamic gas-oil relative permeability correlation that accounts for the compositional changes owing to the produced gas injection is introduced and applied to correct for changes in interfacial tension (IFT), and its effect on oil recovery is examined. The results show that the simple modified black-oil approach can model well both immiscible and miscible floods, as long as the minimum miscibility pressure (MMP) is matched. It provides a fast and robust alternative for large-scale reservoir simulation with the purpose of flaring/venting reduction through reinjecting the produced gas into the reservoir for EOR. Molecular diffusion plays an important role in oil and gas migration in tight shale formations. However, there are insufficient reference data in the literature to specify the diffusion coefficients within porous media. Another objective of this research is to estimate the diffusion coefficients of shale gas, shale condensate, and shale oil at reservoir conditions with CO2 injection for EOR/EGR. The large nano-confinement effects including large gas-oil capillary pressure and critical property shifts could alter the phase behaviors. This study estimates the diffusivities of shale fluids in nanometer-scale shale rock from two perspectives: 1) examining the shift of diffusivity caused by nanopore confinement effects from phase change (phase composition and fluid property) perspective, and 2) calculating the effective diffusion coefficient in porous media by incorporating rock intrinsic properties (porosity and tortuosity factor). The tortuosity is obtained by using tortuosity-porosity relations as well as the measured tortuosity of shale from 3D imaging techniques. The results indicated that nano-confinement effects could affect the diffusion coefficient through altering the phase properties, such as phase compositions and densities. Compared to bulk phase diffusivity, the effective diffusion coefficient in porous shale rock is reduced by 102 to 104 times as porosity decreases from 0.1 to 0.03. Finally, a fully compositional model is developed, which enables us to process multi-component multi-phase fluid flow in shale nano-porous media. The validation results for primary depletion, water injection, and gas injection show a good match with the results of a commercial software (CMG, GEM). The nano-confinement effects (capillary pressure effect and critical property shifts) are incorporated in the flash calculation and flow equations, and their effects on Bakken oil production and Marcellus shale gas production are examined. The results show that including oil-gas capillary pressure effect could increase the oil production but decrease the gas production. Inclusion of critical property shift could increase the oil production but decrease the gas production very slightly. The effect of molecular diffusion on Bakken oil and Marcellus shale gas production are also examined. The effect of diffusion coefficient calculated by using Sigmund correlation is negligible on the production from both Bakken oil and Marcellus shale gas huff-n-puff. Noticeable increase in oil and gas production happens only after the diffusion coefficient is multiplied by 10 or 100 times. / Doctor of Philosophy / Shale reservoir is one type of unconventional reservoir and it has extremely small pore size, low porosity, and ultra-low permeability. In tight shale reservoirs, the pore size is in nanometer scale and the oil-gas capillary pressure reaches hundreds of psi. In addition, the critical properties (such as critical pressure and critical temperature) of hydrocarbon components will be altered in those nano-sized pores. In this research, two in-house reservoir simulation models, i.e., a compositionally extended black-oil model and a fully composition model are developed to examine the nano-pore confinement effects on convective and diffusive multicomponent multiphase fluid transport. The large nano-confinement effects (large gas-oil capillary pressure and critical property shifts) on oil or gas production behaviors will be investigated. Meanwhile, the nano-confinement effects and rock intrinsic properties (porosity and tortuosity factor) on predicting effective diffusion coefficient are also studied.

Actuellement, le module de puissance (convertisseur de puissance) standard hybride 2D est la technologie de référence qui domine le marché de la moyenne et de la forte puissance. Ce dernier se présente sous la forme d'un boitier à multi-puces discrètes. Les puces à semi-conducteur sont reliées entre elles par des faisceaux de wire-bonding (câblage par fils) pour former des cellules de commutation. La technologie d'interconnexion wire-bonding présente une grande maturité technologique, et ses modes de défaillance sont bien connus aujourd'hui. Toutefois, cette technologie est un facteur limitant en termes de performances électrique et thermomécanique, d'intégrabilité tridimensionnelle et de productivité. Ces travaux de thèse ont pour objectif de proposer et d'étudier de nouvelles architectures de convertisseurs de puissance très intégrés. Comparée à la technologie hybride, dite de référence, les architectures proposées visent à un degré d'intégration plus poussé, avec un effort d'intégration partagé et conjoint au niveau semi-conducteur (intégration monolithique) et au niveau assemblage (intégration hybride). L'intégration monolithique consiste à intégrer les interrupteurs formant les cellules de commutation dans de nouvelles architectures de puces, passant ainsi de la notion de puce dipôle à celle de macro-puce multi-pôle. L'intégration hybride repose sur le développement de nouvelles technologies de report et d'assemblage de ces macro-puces. Pour valider les trois nouvelles architectures d'intégrations proposées, la démarche a consisté dans un premier temps à étudier et valider le fonctionnement des nouvelles puces par des simulations SentaurusTM TCAD. Ensuite, les puces multi-pôles ont été réalisées en s'appuyant sur la filière IGBT disponible dans la plateforme de micro-fabrication du LAAS-CNRS. Pour finir, les puces ont été reportées sur des cartes PCB, afin de réaliser des circuits de conversions prototypes. La maille de commutation très intégrée proposée présente une inductance parasite inférieure au nanohenry, ce qui est remarquable comparée à ce qui est présenté dans l'état de l'art (env. 20 nH). / Currently, the standard 2D hybrid power module (power converter) is the reference technology for the medium and high power market. This hybrid power module is a discrete multi-chip case. The semi-conductor chips are interconnected by wire-bonding to form switching cells. The wire-bonding interconnection technology is a limiting factor in terms of electrical and thermomechanical performances, three-dimensional integrability and productivity. The aim of this thesis is to study new architectures of very integrated power converters. Compared to the so-called hybrid reference technology, the proposed architectures aim at a greater degree of integration, with an integration at both the semi-conductor level (monolithic integration) and the packaging level (hybrid integration). Monolithic integration consists in integrating switching cells into new multi-terminal macro-chip architectures. Hybrid integration consists in developing of new technologies to assemble these macro-chips. To validate the different proposed integration architectures, the first step was to study and validate the operating modes of the new chips by SentaurusTM TCAD simulations. Then, the multi-terminal chips were realized in the micro and nanotechnology platform of LAAS-CNRS laboratory. Finally, the chips were bonded on PCB substrates to realize power converter circuit prototypes. The highly integrated switching loop presents a stray inductance loop lower than one nanohenry, wich is an important improvement as compared to the values reported in literature (about 20 nH).

Chloride-induced corrosion of reinforcing steel in concrete is a worldwide problem. In order to predict how chlorides penetrate in concrete and how other ionic species in con-crete pore solution affect the penetration of chlorides, this thesis presents a numerical study on multi-phase modelling of ionic transport in concrete dominated by migration process. There are many advantages in rapid chloride migration test (RCM) method and numeri-cal approach. However, most of models in the literature predicting chloride diffusivity in concrete are diffusion models, which not consider the action of externally applied electric field. In view of this, the specific aim of this thesis is to develop a rational nu-merical migration model to simulate chloride migration tests. By using this model, the diffusion coefficient of chlorides in concrete will be efficiently predicted. Furthermore, other mechanisms of ionic transportation in composite materials can be scientifically in-vestigated in the meantime. In most existing work, researchers tend to use the assumption of electro-neutrality con-dition, which ensures that no external charge can be imported (Bockris and Reddy, 1998), to determine the electrostatic potential within concrete as well as considering a 1-D problem with only one phase structure and single species (i.e. the chlorides) for pre-dicting the ionic migration. In contrast, this thesis presents a number of sets of multi-phase migration models in more than one dimension and uses the Poisson’s equation for controlling the multi-species interactions. By solving both mass conservation and Pois-son’s equations, the distribution profiles of each ionic species and electrostatic potential at any required time are successfully obtained. Some significant factors, i.e. the influ-ence of dimensions, aggregates, interfacial transition zones (ITZs), cracks and binding effect have also been discussed in detail. The results reveal a series of important features which may not be seen from existing numerical models. For quantitative study, this thesis also provides the prediction method of chloride diffu-sivity not only by the traditional stationary diffusion models but also by the migration models presented in the thesis. The obtained results are compared with three proven analytical models, i.e., Maxwell’s model (Dormieux and Lemarchand, 2000), Brug-geman’s equation (Bruggeman’s, 1935) and the lower bound of the effective diffusion coefficient proposed by Li et al. (2012) as well as validated against experimental data sets of an accelerated chloride migration test (ACMT) brought by Yang and Su (2002).

Boost dc-to-dc converters have very good source interface properties. The input inductor makes the source current smooth and hence these converters provide very good EMI performance. On account of this good property, the boost converter is also the preferred converter for off-line UPF rectifiers. One of the issues of concern in these converters is the large size of the storage capacitor on the dc link. The boost converter suffers from the disadvantage of discontinuous current injected to the load. The size of the capacitor is therefore large. Further, the ripple current in the capacitor is as much as the load current; hence the ESR specification of the tank capacitor is quite demanding. This is specially so in the emerging application areas of automotive power conversion, where the input voltage is low (typically 12V) and large voltage boost (4 to 5) are desired. The first part of this thesis suggests multi-phase boost converter to overcome the disadvantages of large size storage capacitor in boost converter. Comparison between the specification of single stage and multi-stages is thoroughly examined. Besides the average small signal analysis of N converters in parallel and obtaining an equivalent second order system are discussed. By paralleling the converters the design of closed loop control is a demanding task. To achieve proper current sharing among the stages using current control method is inevitable. Design and implementation of closed loop control of multi-phase boost converter both in analog and digital is the topic of next part of the thesis. Comparison between these two approaches is presented in this part and it will be shown that digital control is more convenient for such a topology on account of the requirement of synchronization, phase shifted operation, current balancing and other desired functions, which will be discussed later in detail. A new direct digital control method, which is simple and fast, is developed. Two different realizations with DSP controller and FPGA controller are considered. In the last part of the thesis a novel soft switching circuit for boost converter is presented. It provides Zero Voltage Switching (ZVS) for the main switch and Zero Current Switching (ZCS) for the auxiliary switch. The paper presents the idealized analysis giving all the circuit intervals and the equations necessary for the design of such a circuit. The proposed soft switching circuit is particularly suited for the tapped-inductor boost circuit with a minimum number of extra components. Extension of the method to tapped inductor boost converter addresses the application of Zero Voltage Transition (ZVT) to high conversion ratio converters. Extension of the method to multiphase boost converter shows that with less number of auxiliary switches soft switching operation can be achieved for all interleaved switching devices. Several laboratory prototype boost converters have been built to confirm the theoretical results and design methods are matching with both simulation and experimental results.