Title Promoter Affiliations Abstract "Nonlinear Characterization and Modeling of Magnetic Tunnel Junction (MTJ)-Based Magnetic Sensors" "Dominique Schreurs" "ESAT- TELEMIC, Telecommunications and Microwaves" "A magnetic tunnel junction (MTJ) consists of the free layer (FL), pinned layer (PL), and the antiferromagnetic (AFM) layer. The AFM layer fixes the PL’s magnetization orientation via exchange coupling. The FL and PL are separated by a thin, crystalline MgO barrier. Electrons tunnel through the MgO insulating layer resulting in electrical conductivity, which depends on the relative orientation between the magnetizations of the FL and PL. The PL is assumed to be perfectly fixed while the FL is free to move in response to the external magnetic field. Even though it is assumed that the PL is fixed, it is in fact pinned by a finite energy. Room-temperature thermal magnetization fluctuations (i.e. linear magnetization dynamics) of both the FL and PL are the main source of magnetic noise and can be quantified by their ferromagnetic resonance (FMR) modes. Another source of magnetization dynamics in MTJs is the current-induced spin-torque effect, which describes a direct transfer of angular momentum from the spin-polarized electrons to the local magnetization. This effect influences the thermal magnetization fluctuations, and its contribution can be observed directly in the measured FMR spectrum. MTJs also exhibit nonlinear characteristics—nonlinear bias voltage dependence of the tunneling magnetoresistance (TMR) and spin-torque effects. The interaction between the MTJs’ linear magnetization dynamics and nonlinear characteristics results in nonlinear magnetization dynamics. The latter can be quantified by the sub-harmonics of the FL and PL FMR modes as well as the “splitting mode” located above the frequency of the FL. The resultant nonlinear magnetization dynamics are interpreted via micromagnetic modeling. The author provides physical explanations of the observed nonlinear effects and elaborates on their applications. This work will be continued in the course of postdoctoral research." "Continuous learning" "Tinne Tuytelaars" "ESAT - PSI, Processing Speech and Images" "The world around us is continuously evolving (think e.g. of fashion, news items or social media trends). A static artificial agent, trained in the lab and then deployed in the real world, will quickly get outdated. To keep up the pace, an artificial agent should be evolving as well, gradually increasing its knowledge about the world and expanding its horizons. In this project, we study how to make an artificial agent learn multiple tasks, while keeping its memory footprint constant. This involves retaining learned knowledge, adapting to particular working conditions, and improving its internal representation (embedding) along the way. In particular, we will focus on visual image understanding, including complex tasks such as multiple object interactions (triples of ) ." "The Analysis of Thermal Hydraulic Phenomena in the MYRRHA Heavy Liquid-Metal Pool-type Reactor in Forced and Natural Circulation" "William D'haeseleer" "Applied Mechanics and Energy Conversion Section" "SCK•CEN, the Belgian Nuclear Research Centre, is at the forefront of Heavy Liquid Metal nuclear technology worldwide with the development of the MYRRHA Accelerator Driven System. MYRRHA is an innovative flexible fast-spectrum pool-type research reactor cooled by Lead Bismuth Eutectic.The use of passive systems to achieve safety functions is one of the guiding principles of the MYRRHA design and safety approach. Natural circulation in the primary system is the chosen option in terms of decay heat removal. The demonstration that this is viable, is indispensable.The understanding of the thermal hydraulic phenomena occurring in the upper and lower plena of the reactor pool is a critical issue in the design process. A good knowledge of convection patterns, thermal mixing and stratification in operational and accidental conditions is absolutely necessary.The objective of the PhD is to investigate the thermal hydraulic behavior of a heavy-liquid metal pool-type reactor such as MYRRHA, on the basis of experiments in a scaled liquid metal facility." "Value of Flexibility in Future Electricity Markets" "Ronnie Belmans" "ESAT - ELECTA, Electrical Energy and Computer Architectures, Applied Mechanics and Energy Conversion Section" "ContextFlexibility is defined as the ability of power systems to cope with expected and unexpected changes in demand and generation, as well as grid disturbances. New technology trends such as variable, renewable generation and distributed energy resources challenge the demand as well the supply of flexibility, historically provided by conventional power plants. The development and deployment of additional resources for flexibility is therefore assessed as a key issue in facilitating the transition towards a sustainable power system. Besides conventional power plants, well-known sources of flexibility are storage and demand response. In addition, transmission network and market integration facilitate the access to these flexibility resources in other regions.A market place, which correctly prices flexibility, is expected to be crucial in order to provide correct incentives for investments in flexibility. This market needs to match the needs of the system, determined by means of different energy, network and reliability services in the current market design, with the potential of new and existing flexibility providers. As the distribution system level is increasing in importance due to the increase of distributed energy resources, it is advised to research the potential of providing these resources with access to the market for flexibility.  This can be done by means of aggregation in clusters or virtual power plants. It remains however unclear how this market for flexibility will evolve. Will this gradually evolve towards an extension of the current energy and ancillary service markets, or will this give birth to a separate “flexibility market”, trading capacity characterised with a certain response time and activation limits? Will this give rise towards standardized products, improving market liquidity, possibly supported by a dedicated market platform?Research objectiveThe objective of this research proposal is to develop the framework for a future market for flexibility. A model-based approach is pursued to match the demand and supply of flexibility, assessing the need for a product standardization, and the interaction with energy and ancillary markets in the electricity sector.MethodologyFirst of all, this requires a study of the techno-economic potential of different flexibility providers. The core of the first research step is to determine techno-economic potential of different flexibility providers. In addition to existing flexibility providers, specific attention is directed towards the aggregation of distributed energy resources.  In a second step, the needs of the system are studied and the current market products facilitating flexibility are analysed. On the other hand, it is necessary to study the needs of the system. In a third step, it is to be researched how the needs of the system, and the potential of various flexibility providers, can be matched in a market place. It needs to be researched if a certain level of standardization can be achieved, in order to increase market liquidity.A model is built to bring together the demand and supply of flexibility. This model-based approach allows to assess flexibility market design. The model is to be particularly designed to simulate the operation of flexible technologies while participating in multiple flexibility services. This model can be used to assess the techno-economic feasibility of standardized flexibility products, as well as put forward optimal product characteristics. " "Co-design of Mm-wave Interfaces, Duplexers and CMOS Transmitters" "Patrick Reynaert" "ESAT - MICAS, Microelectronics and Sensors" "The vision for both the 5th generation mobile communication systems (5G) and the internet of things (IoT) keeps moving towards the final deployment. These new emerging applications require enormous data transmission with low latency among user devices and different networks. Spectral resources at low-GHz range are running low and could hardly meet the requirement either in a wireless or wire-line fashion. Millimeter-wave (mm-Wave) technology is widely considered as one of the key technologies that will continue to serve the consumer demand for increased wireless data capacity. Meanwhile, the advanced CMOS can now well operate in mm-Wave bands, enabling the integration of a full transceiver in a low-cost, high-yield technology. However, the design of a mm-Wave system in advanced CMOS technologies still poses many challenges at both device and architecture levels. In addition to generic difficulties, such as low active gain, low supply voltage and high parasitic loss, we must deal with the distributed effect caused by smaller wavelength at mm-Wave. The dimension of on-chip passive components becomes comparable with the wavelengths at high frequencies and thus microwave theory starts to dominate the matching network design. More importantly, as the extension of the back end of line (BEOL), the interface of integrated circuits (ICs) and the following passive components are the key responsibles for system performance at mm-Wave. However, present design methodology draw a clear line between on-chip and off-chip module designs. When design separately the optimal performance can be achieved within the block level. However, at mm-Wave the interactions between on-chip and off-chip elements become so strong that such a boundary gradually vanishes with increased frequency. Since the distributed effect exists both on-chip and off-chip, superior system performance can be expected if a co-design can be implemented among them with the help of microwave theory. In this doctoral work, the main goal is to innovate a co-design methodology between mm-Wave CMOS building blocks and their RF interfaces that demonstrate better system performance, higher integration level, and low cost than traditional architectures.The classical circuit and field theories will be reviewed, which together with the transmission line theory, build up the theoretical foundation of the co-design methodology. Typical transmission line formats will be discussed in detail, followed by the fundamentals of impedance matching techniques. These will be extensively practiced in the mm-Wave designs operating at various frequencies. Based on the specific application requirement, details of design consideration, proposed concept, simulation and measurement will be presented. The design examples include: a) An E-band packaging solution based on wirebonding and flip-chip interconnect. Using all standard components and fabrication process, a robust and wideband interconnect from RF GSG bondpad to WR-12 waveguide interconnect is demonstrated. b) A 28 GHz phased array frontend design in 40nm CMOS. The designed array consists of a high-efficiency and linear power amplifier, a 5-bit passive phase shifter, and a 4-by-4 slot-coupled patch antenna array. The designed array achieves 44 dBm peak EIRP and 60 degrees scanning range. c) Packaging and duplex designs at 120 GHz for high-speed communications through dielectric waveguides. The first real time 10 + 10 Gbps duplex communication is demonstrated in measurement. d) A complete dielectric waveguide data link design at 140 GHz. An automatic compensation loop is proposed and verified in measurement and the designed link supports 4FSK modulation, duplex functionality, and enhanced robustness against PVT variations." "The Value of Large-Scale Offshore Energy Storage in the Future Electricity Market" "Ronnie Belmans" "ESAT - ELECTA, Electrical Energy and Computer Architectures" "Electricity is a real-time product. Supply and demand, or generation and consumption, have to match exactly at any instance to support the stable operation of the power system. This results from the fact that electricity is not economically storable on a large scale. While this is challenged by techno-economic developments leading to increased and improved storage capacities, storage levels remain well below that of other energy commodities. In addition, in light of the growing importance of sustainability, there is an ongoing transition towards variable renewable energy sources. Their limited controllability and predictability result in an increasing need for flexibility, which is the ability to provide power adjustments to compensate for temporary imbalances between generation and consumption. At the same time, the flexibility offered by the generation side is threatened by closure of conventional power plants that are currently experiencing decreasing profitability. While flexibility can also be provided by flexible supply, flexible demand, and the electric grid, electricity storage is expected to play an important role to fill the flexibility gap.This thesis studies the participation and modeling, and role and value, of electricity storage in short-term electricity markets, including day-ahead and intra-day energy markets, and real-time balancing markets. These markets are important tools to deal with the variability in the system, in which the need for flexibility is expressed and its provision is valorized. As such, they are becoming increasingly important with the ongoing integration of variable renewables. The geographical scope includes Belgium and the Central Western European region, including the French, German, and Dutch market zones next to that of Belgium.First, the concept of electricity storage is discussed, along with a quantitative study on the role, value, and benefits of storage in the transition to, and operation of, highly renewable power systems. The former includes a discussion on the definition of electricity storage, applications for which storage systems can be used, techno-economic parameters by which storage systems can be characterized, and storage technologies that are often considered for grid integration. The latter includes the presentation of a system-wide generation expansion planning model that decides on the cost-minimizing generation mix and scheduling to meet the demand for energy and frequency control, subject to detailed operating requirements and constraints. This model is applied to a test system to derive system-independent and broadly-applicable insights on the role and value of storage, and the interdependency of flexibility sources.Second, since understanding short-term markets is essential for analyses related to flexibility, their design rules are studied in detail along with the implications for flexibility. This is done for the four market zones of the Central Western European region, and provides insight in whether flexibility is treated consistently and appropriately among the different markets, both in time and in space. Where appropriate, desirable future market reforms are indicated. Third, the storage participation, including its trading and operation, in short-term markets is studied. In a first study, employing storage systems for a single application is considered, namely day-ahead market arbitrage. A single-player storage operator perspective is assumed, resulting in a price-based unit commitment formulation. Detailed operating constraints are considered, and a new methodology to study the price-effect of storage actions is introduced based on so-called market resilience functions. This price-effect states that storage generally reduces price spreads by increasing low prices and decreasing high prices. In addition, a stepwise approximation to the piecewise linear market resilience functions is proposed, offering the capability to reduce computation time while providing accurate lower and upper bounds. The developed models are applied to Belgian market data to quantify the arbitrage value and price-effect. Since determining the true value of storage requires the aggregation of applications, and the co-optimization of these applications to avoid conflicting uses, in a second study the day-ahead market arbitrage models are extended to allow for the aggregation of different arbitrage opportunities in the three short-term markets. In addition, the price-effect is studied for the intra-day and real-time markets as well. These models are used to analyze the opportunities for storage in the three short-term markets and four market zones, while differences in storage value are traced back to differences in market design. Fourth, the aggregation of applications can also be achieved through the co-operation and sharing of storage resources by different players. New markets, or market products within existing markets, to enable such storage uses, and thus the decoupling of ownership from operation, can be valuable levers to capture the true value of storage. The concept of physical storage rights is introduced, which can be auctioned to different players and entitle the holders to the right to use storage resources. Based on a case study with Belgian data, the storage value in a range of fixed a priori allocations is compared to that of allocations resulting from the proposed auction-based mechanism to show its merits." "Large-eddy Simulation and Optimal Coordinated Control of Wind-farm Boundary Layers" "Johan Meyers" "Applied Mechanics and Energy Conversion Section" "In wind farms, velocity deficits in the wakes originating from upstream turbines result in a significant decrease in power extraction in downstream turbine rows. The current control paradigm in industry optimizes power extraction at the turbine level, and does not account for these interactions, leading to sub-optimal wind-farm efficiencies. The current dissertation investigates the use of a coordinated control approach that maximizes power extraction at the wind-farm level instead. To this end, optimal control techniques are combined with a flow model based on large-eddy simulations in which turbines are modeled using an actuator disk approach. Furthermore, sensitivities of wind-farm power extraction to control parameters are evaluated in a tractable manner using an adjoint approach. In this way, individual wind turbines can be employed as dynamic flow actuators to both influence and harness specific flow features of the turbulent atmospheric boundary layer, with the aim of increasing the overall wind-farm power extraction. Although this approach is computationally infeasible for real-time wind-farm control, it allows to benchmark the potential of coordinated control. The current work contains cases studies involving both dynamic axial induction control and dynamic yaw control. It is found that, for all optimal control cases, wind-farm power extraction can be increased significantly. Furthermore, analysis of control dynamics and understanding of the optimized flow physics lead to the derivation of simplified induction and yaw control strategies that could be applied in practice. In addition to the wind-farm control studies, the current work also touches upon the generation of turbulent inflow conditions." "Advanced Modeling of Large Radius Air Bending" "Joost Duflou" "Centre for Industrial Management / Traffic & Infrastructure, Production Engineering, Machine Design and Automation (PMA) Section" "Air bending remains one of the most popular forming techniques within the sheet metal manufacturing domain. This forming process is known for its flexibility since one pair of tooling suffices for a wide range of forming angles. The traditional approach of conventional or small radius air bending is described by a 3-point bending model. However, small radius bending is not technically feasible for high strength steels. High strength steels have superior load bearing characteristics, however, this quality usually results in very limited ductility. In order to be able to form high strength steels, a reduced degree of deformation should be imposed during the forming stage, for instance by using larger radius punches. However, large radius bending brings about several difficulties, such as the so-called multi-breakage effect. This effect makes the prediction of the bending process more difficult and dissimilar to conventional air bending. The objective of this research is to investigate the effect of process and tool parameters in a large radius bending process using both physical and simulation models.The first step in this doctoral project, is an extensive experimental campaign, which is set up to refine the limits of the process window of large radius bending and to establish a cause-effect map on a number of process settings and quality indicators of the formed component. The experimental investigation contains more than 1000 bending tests.Two regression models are constructed, based on the phenomenological observations and on a circular approximation of the bent plate. This approach predicts the bending characteristics accurately, but it requires the availability of a significant experimental data set.The simulation part of the work is performed using finite element models which turn out to be quite accurate. Both mesh parameters and relevant material parameters are set to achieve close correlation with experiments.An analytical model predicts the bending characteristics based on a circular approximation approach. Only a limited number of input parameters is required for this model. The prediction quality of the analytical model is comparable with the regression approach and it is better than finite element analysis predictions. However, it does not require an extensive database as for regression analysis, or significant computational resources and commercial calculation tools as for the numerical analysis.In the final step, the possibilities of process window enlargement for air bending by radiant heating are addressed. Forming of thick plates with a small radius punch is possible while simultaneously decreasing the springback and required bending force." "Fully-Integrated Wide Input Range and Two-Quadrant Switched-Capacitor DC-DC Converters" "Michiel Steyaert" "ESAT - MICAS, Microelectronics and Sensors" "This work focuses on two specific applications of fully-integrated switched-capacitor DC-DC converters, namely wide range operation and two-quadrant operation. For switched-capacitor DC-DC converters to operate efficiently over a wide input or output voltage range, multiple Voltage Conversion Ratios (VCRs) are required, to counteract the losses associated to its output impedance. To achieve wide range operation, this work proposes to use the Dickson topology as its foundation, transforming it into a wide range topology called the folding Dickson topology. Here, multiple VCRs are achieved through changing the phases of the switches. This operation lumps two or more flying capacitors into a single flying capacitor, which allows the emulation of new VCRs. Furthermore, a small modification to the output stage increases the amount of VCRs even more. These findings were validated through silicon measurements, and boasted an average efficiency of 71% over an input voltage range from 2.5V up to 8V, which is more than 4.5 times the technologies supply voltage. The regularity of this topology allows it to be extended to achieve even higher voltage ranges, at a minimal design overhead.Two-quadrant DC-DC converters on the other hand might alleviate issues that are encountered with getting the ever increasing currents on-chip. In stead of supplying loads in parallel, they are put in stacked voltage domains. The shared node now needs to be balanced by a DC-DC converter, capable of both sourcing and sinking current. Switched-capacitor DC-DC converters require two VCRs to achieve both modes of operation. This work presented the first fully-integrated two-quadrant switched-capacitor DC-DC converter, capable of supplying loads in stacked voltage domains with variable load voltages (0.75-1.1V), while very high systems efficiencies up to 99.6% are achievable thanks to the low-power two-fold control loop." "Vertical Transistors: a slippery path towards the ultimate CMOS scaling" "Kristin De Meyer" "Associated Section of ESAT - INSYS, Integrated Systems" "The semiconductor industry has largely relied on Moore’s law, based on the observation that every new generation of transistors has been better than the previous one in Power, Performance, Area and Cost (PPAC) metrics simultaneously. However, this trend is under a pressure now. The main issue is related to the enormous complexity of both technology and design, which drastically raises not only the manufacturing, but also the R&D costs. Therefore, in order to minimize risks and maximize benefits of a new technology, it is being co-optimized hand in hand with a design relying on this technology.The scaling of lateral transistors is going to reach its limit soon because it mainly relies on the scaling of contacted gate pitch (CGP), which, in turn, forces the scaling of gate length, S/D spacers and contacts. Reduction of any of these dimensions is undesirable as it leads to poorer electrostatic control, increased parasitic capacitance and increased access resistance, respectively. There are lateral devices, like nanowire-based FETs, which may postpone the problem of CGP budgeting but they cannot solve it.The focus of this PhD work is on the vertical devices. These devices are less constrained on gate length and spacer thickness as they are oriented vertically and thus should demonstrate better scalability than lateral transistors. We quantify the advantages of the vertical devices in terms of PPA metrics through a holistic benchmark by combining the design techniques and technology limitations which are likely to be in place at the 5nm technology. In order to do this, we perform the layouts analysis, model and evaluate RC parasitics, calibrate compact models to TCAD and experimental data. Afterwards, we run simulations on a ring oscillator level to extract the PPA metrics.We have not limited ourselves to the conventional MOSFETs only, but we also benchmark vertical III-V heterojunction Tunnel FETs in order to get a better understanding under which conditions the vertical architecture is the most advantageous. This allows us to shed light on the ultimate CMOS scaling and to understand whether the introduction of vertical transistors can enable the next technological nodes."