No description available.

No description available.

No description available.

No description available.

This thesis mainly addresses the problems of thermal management in data centers (DCs) through data-driven techniques. For thermal management, a temperature prediction model in the facility is very important, while the thermal modeling based on first principles in DCs is quite difficult due to the complicated air flow and heat transfer. Therefore, we employ multiple data-driven techniques including statistical methods and deep neural networks (DNNs) to represent the thermal dynamics. Then based on such data-driven models, temperature estimation and control are implemented to optimize the thermal management in DCs. The contributions of this study are summarized in the following four aspects: 1) A data-driven model constructed through multiple linear Autoregression exogenous (ARX) models is adopted to describe the thermal behaviors in DCs. On the basis of such data-driven model, an observer of adaptive Kalman filter is proposed to estimate the temperature distribution in DC. 2) Based on the data-driven model proposed in the first work, a data-driven fault tolerant predictive controller considering different actuator faults is developed to regulate the temperature in DC. 3) To improve the modeling accuracy, a deep input convex neural network (ICNN) is adopted to implement thermal modeling in DCs, which is also specifically designed for further control design. Besides, the algorithm of elastic weight consolidation (EWC) is employed to overcome the catastrophic forgetting in continual learning. 4) A novel example reweighting algorithm is utilized to enhance the robustness of ICNN against noisy data and avoid overfitting in the training process. Finally, all the proposed approaches are validated in real experiments or experimental-data-based simulations. / Dissertation / Doctor of Philosophy (PhD) / This thesis mainly investigates the applications of data-driven techniques for thermal management in data centers. The implementations of thermal modeling, temperature estimation and temperature control in data centers are the key contributions in this work. First, we design a data-driven statistical model to describe the complicated thermal dynamics of data center. Then based on the data-driven model, efficient observer and controller are developed respectively to optimize the thermal management in data centers. Moreover, to improve the nonlinear modeling performance in data centers, specific deep input convex neural networks capable of good representation capability and control tractability are adopted. This thesis also proposes two novel strategies to avoid the influence of catastrophic forgetting and noisy data respectively during the training processes. Finally, all the proposed techniques are validated in real experiments or experimental-data-based simulations.

The ever-growing demand for cyber-physical infrastructures has significantly affected worldwide energy consumption and environmental sustainability over the past two decades. Although the average heat load of the computing infrastructures has increased, the supportive capacity of cooling infrastructures requires further improvement. Consequently, energy-efficient cooling architectures, real-time load management, and waste heat utilization strategies have gained attention in the data center (DC) industry. In this dissertation, essential aspects of cooling system modularization, workload management, and waste-heat utilization were addressed. At first, benefits of several legacy and modular DCs were assessed from the viewpoint of the first and second laws of thermodynamics. A computational fluid dynamics simulation-informed thermodynamic energy-exergy formulation captured equipment-level inefficiencies for various cooling architectures and scenarios. Furthermore, underlying reasons and possible strategies to reduce dominant exergy loss components were suggested. Subsequently, strategies to manage cooling parameters and IT workload were developed for the DCs with rack-based and row-based cooling systems. The goal of these management schemes was to fulfill either single or multiple objectives such as energy, exergy, and computing efficiencies. Thermal models coupled to optimization problems revealed the non-trivial tradeoffs across various objective functions and operation parameters. Furthermore, the scalability of the proposed approach for a larger DC was demonstrated. Finally, a waste heat management strategy was developed for new-age infrastructures containing both air- and liquid-cooled servers, one of the critical issues in the DC industry. Exhaust hot water from liquid-cooled servers was used to drive an adsorption chiller, which in turn produced chilled water required for the air-handler units of the air-cooled system. This strategy significantly reduced the energy consumption of existing compression chillers. Furthermore, economic and environmental assessments were performed to discuss the feasibility of this solution for the DC community. The work also investigated the potential tradeoffs between waste heat recovery and computing efficiencies. / Thesis / Doctor of Philosophy (PhD)

The mass-market adoption of electrification in the transportation sector mandates stringent and aggressive requirements in terms of cost, power rating, efficiency, power density, and specific density of power electronics. Modular packaging of power electronics is advantageous and thus ubiquitously used by the automotive industry. A trend of shrinking die sizes and increased integration is evident and will inevitably continue. The thermal management system has become ever more significant as it is one of the main obstacles to higher power densities. The cooling system must be cost-effective, simple, efficient, reliable, and compatible with system requirements. Pin fins are a reliable and effective means of augmenting heat transfer. They rely on inducing turbulence, increasing the effective wetted surface, and accelerating fluid velocity. Unavoidably the pin fin array also produces an undesirable pressure drop that is commensurate to the pumping power required for the system. In this thesis, a tool is developed for the geometry optimization of pin fin arrays to dissipate the heat at a rate large enough to ensure junction temperatures do not exceed the maximum value possible at a minimal pressure drop. It is hoped that this tool would contribute to the multi-physics optimization and integration of power electronics for electrified vehicles. This optimization is confined to equalaterally spaced short pin fins, aspect ratios less than three. The tool employs empirical correlations since flow is too complex to solve analytically and numerical solutions or CFD-simulations are too time and computationally extensive. The tool development is done in a comprehensive manner. Starting from the first principles of a two-level voltage source inverter's operation. Next, the inevitable power losses from the operation are explained and a method for their calculations is presented. Correlations in the literature related to both pressure drop and heat transfer are reviewed afterward. Then the methodology of the construction of the tool is explicated in detail. Employing a commercial power module to benchmark results; three scenarios with different flow rates and inlet temperatures are optimized for. Simulations in ANSYS Fluent are run to verify the accuracy of correlations used in the tool. Comparing the optimized geometry of pin fins to the original benchmarking geometry it is evident that employing this tool on a per-application basis provides superior performance. / Thesis / Master of Applied Science (MASc)

This thesis consists of three main objectives: fundamental and literature review of EV batteries, experimental development, and validation of two liquid cooling battery modules, thermal modeling and comparison of the inter-cell cooling battery module. / The traditional vehicles with internal combustion engine have resulted in severe environmental pollution, which motivates the development of electric vehicles and hybrid electric vehicles. Due to a low energy density and long refueling time of the battery pack, it is still hard for electric vehicles and hybrid electric vehicles to be widely accepted by the consumers. As the batteries with a better ultra-fast charging capability are massively produced, the range anxiety issue is somewhat alleviated. During a charging with large current magnitude, the battery generally has a great amount of heat generation and evident temperature rise. Therefore, a thermal management system is necessary to effectively dissipate the battery loss and minimize the degradation mechanisms caused by extreme temperature. The motivation of this thesis is to study the discipline of the battery thermal management system as an application for electric vehicles. The design methodologies are presented in both experiment test and numerical simulation. For the comparative study between active liquid cooling methods for a lithium-ion battery module using experimental techniques, two battery modules with three Kokam Nickel Manganese Cobalt battery cells connected in parallel are developed. One has liquid coolant flowing along the edge of the model, and another with liquid coolant flowing between the cells. Several characterization tests, including thermal resistance tests, fast charging tests up to 5C, and drive cycle tests are designed and performed on the battery module. The inter-cell cooling module has a lower peak temperature rise and faster thermal response compared to the edge cooling module, i.e., 4.1⁰C peak temperature rise under 5C charging for inter-cell cooling method and 14.2⁰C for edge cooling method. The thermal models built in ANSYS represent the numerical simulation of the inter-cell cooling module as a comparison with the experiment. A cell loss model is developed to calculate the battery heat generation rate under ultra-fast charging tests and a road trip test, which are further adopted as the inputs to the thermal models. The simulation of the 5C ultra-fast charging test gives the peak temperature rise just 0.47⁰C lower than the experimental measurement, it indicates that the FEA thermal models can provide an accurate temperature prediction of the battery module. / Thesis / Master of Applied Science (MASc) / With a demanding market of electric vehicles, battery technologies have grown rapidly in recent years. Among all the battery research topics, the development of ultra-fast charging, that can fully charge the battery pack within 15 minutes, is the most promising direction to address the range anxiety and improve the social acceptance of electric vehicles. Nevertheless, the application of ultra-fast charging has many challenges. In particular, an efficient thermal management system is significant to guarantee the safety and prolong the service life of the battery pack. This thesis contributes to study the fundamentals of the battery field, and design liquid cooling systems to observe the thermal behavior of a battery prototype module under fast charging and general use. FEA thermal modeling of the battery module is developed to provide a guide for further test validation.

As silicon processes scale, system-on-chips (SoCs) will require numerous hardware monitors that perform assessment of physical characteristics that change during the operation of a device. To address the need for high-speed and coordinated transport of monitor data in a SoC, we develop a new interconnection network for monitors - the monitor network on chip (MNoC). Data collected from the monitors via MNoC is collated by a monitor executive processor (MEP) that controls the operation of the SoC in response to monitor data. In this thesis, we developed the architecture of MNoC and the infrastructure to evaluate its performance and overhead for various network parameters. A system level architectural simulation can then be performed to ensure that the latency and bandwidth provided by MNoC are sufficient to allow the MEP to react in a timely fashion. This typically translates to a system level benefit that can be assessed using architectural simulation. We demonstrate in this thesis, the employment of MNoC for two specific monitoring systems that involve thermal and delay monitors. Results show that MNoC facilitates employment of a thermal-aware dynamic frequency scaling scheme in a multicore processor resulting in improved performance. It also facilitates power and performance savings in a delay -monitored multicore system by enabling a better than worst case voltage and frequency settings for the processor.

京都大学 / 新制・課程博士 / 博士(工学) / 甲第24645号 / 工博第5151号 / 新制||工||1983(附属図書館) / 京都大学大学院工学研究科化学工学専攻 / (主査)准教授 渡邉 哲, 教授 佐野 紀彰, 教授 河瀬 元明 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DGAM