Dissertations / Theses on the topic 'Adaptive power management'

With the rapid growth of servers and applications spurred by the Internet economy, power consumption in today's data centers is reaching unsustainable limits. This has led to an imminent financial, technical and environmental crisis that is impacting the society at large. Hence, it has become critically important that power consumption be efficiently managed in these computing power-houses of today. In this work, we revisit the issue of adaptive power and performance management of data center server platforms. Traditional data center servers are statically configured and always over-provisioned to be able to handle peak load. We transform these statically configured data center servers to clairvoyant entities that can sense changes in the workload and dynamically scale in capacity to adapt to the requirements of the workload. The over-provisioned server capacity is transitioned to low-power states and they remain in those states for as long as the performance remains within given acceptable thresholds. The platform power expenditure is minimized subject to performance constraints. This is formulated as a performance-per-watt optimization problem and solved using analytical power and performance models. Coarse-grained optimizations at the platform-level are refined by local optimizations at the devices-level namely - the processor & memory subsystems. Our adaptive interleaving technique for memory power management yielded about 48.8% (26.7 kJ) energy savings compared to traditional techniques measured at 4.5%. Our adaptive platform power and performance management technique demonstrated 56.25% energy savings for memory-intensive workload, 63.75% savings for processor-intensive workload and 47.5% savings for a mixed workload while maintaining platform performance within given acceptable thresholds.

Internet of Things (IoT), as broadband network connecting every physical objects, is becoming more widely available in various industrial, medical, home and automotive applications. In such network, the physical devices, vehicles, medical assistance, and home appliances among others are supposed to be embedded by sensors, actuators, radio frequency (RF) antennas, memory, and microprocessors, such that these devices are able to exchange data and connect with other devices in the network. Among other IoT’s pillars, wireless sensor network (WSN) is one of the main parts comprising massive clusters of spatially distributed sensor nodes dedicated for sensing and monitoring environmental conditions. The lifetime of a WSN is greatly dependent on the lifetime of the small sensor nodes, which, in turn, is primarily dependent on energy availability within every sensor node. Predominantly, the main energy source for a sensor node is supplied by a small battery attached to it. In a large WSN with massive number of deployed sensor nodes, it becomes a challenge to replace the batteries of every single sensor node especially for sensor nodes deployed in harsh environments. Consequently, powering the sensor nodes becomes a key limiting issue, which poses important challenges for their practicality and cost. Therefore, in this thesis we propose enabling WSN, as the main pillar of IoT, by means of resonant inductive coupling (RIC) wireless power transfer (WPT). In order to enable efficient energy delivery at higher range, high quality factor RIC-WPT system is required in order to boost the magnetic flux generated at the transmitting coil. However, an adaptive front-end is essential for self-tuning the resonant tank against any mismatch in the components values, distance variation, and interference from close metallic objects. Consequently, the purpose of the thesis is to develop and design an adaptive efficient switch-mode front-end for self-tuning in WPT receivers in multiple receiver system. The thesis start by giving background about the IoT system and the technical bottleneck followed by the problem statement and thesis scope. Then, Chapter 2 provides detailed backgrounds about the RIC-WPT system. Specifically, Chapter 2 analyzes the characteristics of different compensation topologies in RIC-WPT followed by the implications of mistuning on efficiency and power transfer capability. Chapter 3 discusses the concept of switch-mode gyrators as a potential candidate for generic variable reactive element synthesis while different potential applications and design cases are provided. Chapter 4 proposes two different self-tuning control for WPT receivers that utilize switch-mode gyrators as variable reactive element synthesis. The performance aspects of control approaches are discussed and evaluated as well in Chapter 4. The development and exploration of more compact front-end for self-tuned WPT receiver is investigated in Chapter 5 by proposing a phase-controlled switched inductor converter. The operation and design details of different switch-mode phase-controlled topologies are given and evaluated in the same chapter. Finally, Chapter 6 provides the conclusions and highlight the contribution of the thesis, in addition to suggesting the related future research topics.
Internet de las cosas (IoT), como red de banda ancha que interconecta cualquier cosa, se está estableciendo como una tecnología valiosa en varias aplicaciones industriales, médicas, domóticas y en el sector del automóvil. En dicha red, los dispositivos físicos, los vehículos, los sistemas de asistencia médica y los electrodomésticos, entre otros, incluyen sensores, actuadores, subsistemas de comunicación, memoria y microprocesadores, de modo que son capaces de intercambiar datos e interconectarse con otros elementos de la red. Entre otros pilares que posibilitan IoT, la red de sensores inalámbricos (WSN), que es una de las partes cruciales del sistema, está formada por un conjunto masivo de nodos de sensado distribuidos espacialmente, y dedicados a sensar y monitorizar las condiciones del contexto de las cosas interconectadas. El tiempo de vida útil de una red WSN depende estrechamente del tiempo de vida de los pequeños nodos sensores, los cuales, a su vez, dependen primordialmente de la disponibilidad de energía en cada nodo sensor. La fuente principal de energía para un nodo sensor suele ser una pequeña batería integrada en él. En una red WSN con muchos nodos y con una alta densidad, es un desafío el reemplazar las baterías de cada nodo sensor, especialmente en entornos hostiles, como puedan ser en escenarios de Industria 4.0. En consecuencia, la alimentación de los nodos sensores constituye uno de los cuellos de botella que limitan un despliegue masivo práctico y de bajo coste. A tenor de estas circunstancias, en esta tesis doctoral se propone habilitar las redes WSN, como pilar principal de sistemas IoT, mediante sistemas de transferencia inalámbrica de energía (WPT) basados en acoplamiento inductivo resonante (RIC). Con objeto de posibilitar el suministro eficiente de energía a mayores distancias, deben aumentarse los factores de calidad de los elementos inductivos resonantes del sistema RIC-WPT, especialmente con el propósito de aumentar el flujo magnético generado por el inductor transmisor de energía y su acoplamiento resonante en recepción. Sin embargo, dotar al cabezal electrónico que gestiona y condicionada el flujo de energía de capacidad adaptativa es esencial para conseguir la autosintonía automática del sistema acoplado y resonante RIC-WPT, que es muy propenso a la desintonía ante desajustes en los parámetros nominales de los componentes, variaciones de distancia entre transmisor y receptores, así como debido a la interferencia de objetos metálicos. Es por tanto el objetivo central de esta tesis doctoral el concebir, proponer, diseñar y validar un sistema de WPT para múltiples receptores que incluya funciones adaptativas de autosintonía mediante circuitos conmutados de alto rendimiento energético, y susceptible de ser integrado en un chip para el condicionamiento de energía en cada receptor de forma miniaturizada y desplegable de forma masiva. La tesis empieza proporcionando una revisión del estado del arte en sistemas de IoT destacando el reto tecnológico de la alimentación energética de los nodos sensores distribuidos y planteando así el foco de la tesis doctoral. El capítulo 2 sigue con una revisión crítica del statu quo de los sistemas de transferencia inalámbrica de energía RIC-WPT. Específicamente, el capítulo 2 analiza las características de diferentes estructuras circuitales de compensación en RIC-WPT seguido de una descripción crítica de las implicaciones de la desintonía en la eficiencia y la capacidad de transferencia energética del sistema. El capítulo 3 propone y explora el concepto de utilizar circuitos conmutados con función de girador como potenciales candidatos para la síntesis de propósito general de elementos reactivos variables sintonizables electrónicamente, incluyendo varias aplicaciones y casos de uso. El capítulo 4 propone dos alternativas para métodos y circuitos de control para la autosintonía de receptores de energía

In order to run and manage resource-intensive high-performance applications, large-scale computing and storage platforms have been evolving rapidly in various domains in both academia and industry. The energy expenditure consumed to operate and maintain these cloud computing infrastructures is a major factor to influence the overall profit and efficiency for most cloud service providers. Moreover, considering the mitigation of environmental damage from excessive carbon dioxide emission, the amount of power consumed by enterprise-scale data centers should be constrained for protection of the environment.Generally speaking, there exists a trade-off between power consumption and application performance in large-scale computing systems and how to balance these two factors has become an important topic for researchers and engineers in cloud and HPC communities. Therefore, minimizing the power usage while satisfying the Service Level Agreements have become one of the most desirable objectives in cloud computing research and implementation. Since the fundamental feature of the cloud computing platform is hosting workloads with a variety of characteristics in a consolidated and on-demand manner, it is demanding to explore the inherent relationship between power usage and machine configurations. Subsequently, with an understanding of these inherent relationships, researchers are able to develop effective power management policies to optimize productivity by balancing power usage and system performance. In this dissertation, we develop an autonomic power-aware system management framework for large-scale computer systems. We propose a series of techniques including coarse-grain power profiling, VM power modelling, power-aware resource auto-configuration and full-system power usage simulator. These techniques help us to understand the characteristics of power consumption of various system components. Based on these techniques, we are able to test various job scheduling strategies and develop resource management approaches to enhance the systems' power efficiency.

The proliferation of wireless ad-hoc networks especially wireless LAN (IEEE 802.11b Standard) in the commercial market in recent years has reached a critical mass. The adoption and strong support of wireless IEEE 802.11 standard, coupled with the consequent decline in costs, has made wireless LAN deployment as one of the fastest growth area in communication access technology. With the ever increasing use of wireless LAN technology the various networks are reaching their full capacity in terms of network throughput, number of users and interference level in the wireless channel. In this thesis work I propose to the use smart antenna technology and a power management scheme in the wireless ad-hoc networks to increase the network capacity in terms of throughput, number of simultaneous communication and to lower the average transmit power and consequently co-channel interference. Power management scheme can be used to maximize the power efficiency of the transmitter by choosing an optimum transmit power level. Smart antenna or adaptive antenna array technology has reached a level of sophistication that it is feasible to use it on small mobile terminals like handheld PDA, LAPTOP and other mobile devices with limited battery power. The simulation results of various ad â hoc network scenario shows that there are significant gains to be had if these technologies can be integrated in the existing wireless LAN physical layer and/or in the standard them self. Smart antennas along with slight modification in channel access scheme reduce co-channel interference dramatically and increases the number of simultaneous transmissions hence improves network throughput. Power management algorithm is shown to improve average transmission of a node. We present a mathematical framework to characterize the outage probability of cellular mobile radio system with selective co-channel interference receiver in overloaded array environments. The mathematical framework outlines a general numerical procedure for computing the probability of outage of a cellular mobile radio system that is equipped with a smart antenna to suppress a few strongest co-channel interferers (CCI) out of a total of NI active interferers by null steering.
Master of Science

The Smartphone revolution and the Internet of Things (IoTs) have triggered rapid advances in complex system-on-chips (SoCs) that increasing provide more functionality within a tight power budget. Highly power efficient on die switched-capacitor voltage regulators suffer from large output voltage ripple preventing their widespread use in modern integrated circuits. With technology scaling and increasing architectural complexity, the number of transistors switching in a power domain is growing rapidly leading to major issues with respect to voltage noise. The large voltage and frequency guard-bands present in current microprocessor designs to combat voltage noise both degrade the performance and erode the energy efficiency of the design. In an effort to reduce guard-bands, adaptive clocking based systems combat the problem of voltage noise by adjusting the clock frequency during a voltage droop to avoid timing failure. This thesis presents an integrated power management and clocking scheme that utilizes clock-data compensation to achieve adaptive clocking. The design is capable of automatically con figuring the supply voltage given a target clock frequency for the load circuit. Furthermore, during a voltage droop the design adjusts clock frequency to meet critical path timing margins while simultaneously increasing the current delivered to the load to recover from the droop. The design was implemented in IBM's 130nm technology and simulation results show that the design is able to clock the load circuit from 30 MHz to 800 Mhz with current efficiencies as high as 97%.

This thesis advances a conceptual model of adaptive impact monitoring that is designed to overcome many of the criticisms plaguing conventional monitoring strategies. The potential for applying the adaptive model is demonstrated for the Peace River Site C dam proposed for northeastern British Columbia. Environmental impact assessment (EIA) has progressed considerably from its early biophysical orientation to a more comprehensive, interdisciplinary process concerned with the breadth of environmental and socio-economic impacts of development. Impact monitoring, an essential EIA component, has also progressed but in a less innovative fashion. Consequently, conventional monitoring strategies often contain significant deficiencies including insufficient use of past experience, poor monitoring design, and failure to recognize the learning opportunity offered by each project. Adaptive impact monitoring offers significant advantages over traditional strategies. An adaptive strategy is based on a series of impact hypotheses established and tested by an interdisciplinary design team and has two fundamental stages: design and evaluation. A review of the potential environmental impacts of hydroelectric production indicates that the reservoir impact paradigm is beginning to provide a comprehensive basis for assessing development effects. Although the Site C EIA adequately reflects the reservoir impact paradigm, it has three significant weaknesses. First, the potential impacts on downstream ecology and distant downstream users are ill-considered. Second, the potential for increased Site C fisheries parasitism is neglected. Finally, estimates of maximum sustainable yield for the Site C reservoir and Peace River fisheries are unreliable. While opportunities for future impact monitoring were recognized through the Site C panel hearings, they lacked flexibility. The potential impacts on downstream water temperature and fisheries resources are used to illustrate the applicability of the adaptive strategy and the advantages derived from collecting only relevant, statistically credible data to permit testing impact hypotheses in a cost-effective manner. On the basis of these findings, six major policy recommendations are provided for improving the effectiveness of impact monitoring and management for future resource developments.
Graduate and Postdoctoral Studies
Graduate

The focus of this dissertation is adaptive CPU-budget allocation for periodic soft-real-time applications. The presented algorithms are developed in the context of a power-management framework. First, the prediction-based bandwidth scheduler (PBS) is developed. This algorithm is designed to adapt CPU-budget allocations at a faster rate than previous adaptive algorithms. Simulation results are presented to demonstrate that this approach allows for a faster response to under allocations than previous algorithms. A second algorithm is presented called Two-Stage Prediction (TSP) that improves on the PBS algorithm. Specifically, a more sophisticated algorithm is used to predict execution times and a stronger guarantee is provided on the timeliness of jobs. Implementation details and experimental results are presented for both the PBS and TSP algorithms. An abstraction is presented called virtual instruction count (VIC) to allow for more efficient budget allocation in power-managed systems. Power management decisions affect job-execution times. VIC is an abstract measure of computation that allows budget allocations to be made independent of power-management decisions. Implementation details and experimental results are presented for a VIC-based budget mechanism. Finally, a power-management framework is presented called the linear adaptive models based system (LAMbS). LAMbS is designed to minimize power consumption while honoring budget allocations specified in terms of VIC.

Energy-efficient radio frequency (RF) power amplifiers (PAs) are critical and paramount to achieve longer battery life in state-of-the-art portable systems because they typically determine and dominate the power consumption of such devices. In this dissertation, a high-efficiency, linear RF PA with a dynamically adaptive supply and bias current control for code division multiple access (CDMA) and wideband CDMA (WCDMA) is conceived, simulated, and experimentally demonstrated with a discrete PCB-level design and in integrated circuit (IC) form. The PA efficiency is improved by dynamically adjusting both its supply voltage and bias current, there by minimizing its quiescent power dissipation. The PA supply voltage is derived from the battery by a noninverting, synchronous buck-boost switching regulator because of its flexible functionality and high efficiency. Adjusting the PA supply voltage and bias current by tracking the output power, instead of following the complete envelope in large baseband bandwidth wireless applications, is achieved by a converter with a lower switching frequency and consequently higher light-load efficiency, which translates to prolonged battery life. A discrete PCB-level prototype of the proposed system with 915 MHz center frequency, CDMA IS-95 signal having 27-dBm peak-output power resulted in more than four times improvement in the average efficiency compared to a fixed-supply class-AB PA while meeting the required performance specifications. In the IC solution fabricated in AMIs 0.5-micron CMOS process through MOSIS, a dual-mode, buck-boost converter with pulse-width modulation (PWM) control for high power and pulse-frequency modulation (PFM) for low power is designed and implemented to improve the PA efficiency during active and standby operation, respectively. The performance of the dynamically adaptive supply and bias control IC was validated by realizing a 25-dBm, 1.96 GHz center frequency, WCDMA PA over an input supply range of 1.4 4.2 V. The PA with dual-mode power supply and bias control IC showed an average-efficiency improvement of seven times compared to a fixed-supply class-AB PA, which translates to five times improvement in battery life assuming the PA is active for 2 % of the total time and in standby mode otherwise.

Battery storage devices have been widely utilized for different applications. However, for high power applications, battery storage systems come with several challenges, such as the thermal issue, low power density, low life span and high cost. Compared with batteries, supercapacitors have a lower energy density but their power density is very high, and they offer higher cyclic life and efficiency even during fast charge and discharge processes. In this dissertation, new techniques for the control and energy management of the hybrid battery-supercapacitor storage system are developed to improve the performance of the system in terms of efficiency, power quality and reliability. To evaluate the findings of this dissertation, a laboratory-scale DC microgrid system is designed and implemented. The developed microgrid utilizes a hybrid lead-acid battery and supercapacitor energy storage system and is loaded under various grid conditions. The developed microgrid has also real-time monitoring, control and energy management capabilities. A new control scheme and real-time energy management algorithm for an actively controlled hybrid DC microgrid is developed to reduce the adverse impacts of pulsed power loads. The developed control scheme is an adaptive current-voltage controller that is based on the moving average measurement technique and an adaptive proportional compensator. Unlike conventional energy control methods, the developed controller has the advantages of controlling both current and voltage of the system. This development is experimentally tested and verified. The results show significant improvements achieved in terms of enhancing the system efficiency, reducing the AC grid voltage drop and mitigating frequency fluctuation. Moreover, a novel event-based protection scheme for a multi-terminal DC power system has been developed and evaluated. In this technique, fault identification and classifications are performed based on the current derivative method and employing an artificial inductive line impedance. The developed scheme does not require high speed communication and synchronization and it transfers much less data when compared with the traditional method such as the differential protection approach. Moreover, this scheme utilizes less measurement equipment since only the DC bus data is required.

Well-designed demand response is expected to play a vital role in operatingpower systems by reducing economic and environmental costs. However,the current system is operated without much information on the benefits ofend-users, especially the small ones, who use electricity. This thesis proposes aframework of operating power systems with demand models including the diversityof end-users’ benefits, namely adaptive load management (ALM). Sincethere are a large number of end-users having different preferences and conditionsin energy consumption, the information on the end-users’ benefits needsto be aggregated at the system level. This leads us to model the system ina multi-layered way, including end-users, load serving entities, and a systemoperator. On the other hand, the information of the end-users’ benefits can beuncertain even to the end-users themselves ahead of time. This information isdiscovered incrementally as the actual consumption approaches and occurs. Forthis reason ALM requires a multi-temporal model of a system operation andend-users’ benefits within. Due to the different levels of uncertainty along thedecision-making time horizons, the risks from the uncertainty of informationon both the system and the end-users need to be managed. The methodologyof ALM is based on Lagrange dual decomposition that utilizes interactive communicationbetween the system, load serving entities, and end-users. We showthat under certain conditions, a power system with a large number of end-userscan balance at its optimum efficiently over the horizon of a day ahead of operationto near real time. Numerical examples include designing ALM for theright types of loads over different time horizons, and balancing a system with a large number of different loads on a congested network. We conclude thatwith the right information exchange by each entity in the system over differenttime horizons, a power system can reach its optimum including a variety ofend-users’ preferences and their values of consuming electricity.

Africa’s quest for economic development will require the increased availability and use of its abundant energy resources. Nevertheless, most of its rural population remains without access to modern energy services and urban residents typically only enjoy an intermittent supply of electricity. The dominant approach to energy planning in West Africa is top-down and centralized, emphasizing electricity generation from large dams or fossil-fueled plants and subsequent grid extension to reach more customers. However, an alternative and complementary paradigm is that of decentralized or Distributed Generation (DG), which stresses small-scale, on-site generation of power and offers a bottom-up approach to energy development. The goal of this dissertation project is to assess the various options for regional electrification and integration through a holistic analysis of the set of existing technologies and policies for deploying them. The main organ of the Economic Community of West African States (ECOWAS) for regional electricity planning is the West African Power Pool (WAPP) and its primary policy document, the “Master Plan”, addresses regional power supply shortage through centralized planning. Both the WAPP policy documents and the majority of the country-level planning documents are considered to be based on a traditional, empiricist, policy analysis that appears to provide value-neutral solutions and generalizations. In contrast, the analysis provided in this project situates itself within the post-positivistic, deliberative and more contextual approach to policy analysis in order to compare the centralized approach to generation with a distributed approach, which is currently marginal in the region. It uses the Adaptive Management (AM) framework for this analysis, particularly because of the way it deals with ecological resilience in the face of widespread uncertainty. The main policy issue that this project seeks to address is the need for an integrated energy-environment planning process, which is currently lacking in West Africa, so as to achieve long term sustainability. Adaptive management offers policy makers a holistic lens with which to view energy policy, but there are very few examples of institutions that have attempted to implement it in practice anywhere in the world. These instances, however, represent a valuable historical reference point for future policy research and management efforts that seek to explore this approach. In alignment with that objective, this dissertation first provides an overview of the concept of adaptive management in general, and its application to energy problems in particular. Secondly, the research project undertakes a policy analysis of the ECOWAS strategy for electrification, based on a stakeholder analysis, a review of life cycle assessments of existing energy technologies, the expected outcomes of the electricity sector, and a set of traditional criteria for evaluating public policies. In order to further examine the question of electricity access, it carries out a quantitative analysis of the electricity demand and supply in the region. It uses a modeling approach that is based on the logic of AM to determine whether or not the energy requirements for broad based electrification can be met through distributed renewable power, which is currently a negligible component of the generation resource portfolio in West Africa. The dissertation proceeds to carry out a retrospective analysis of three cases in the U.S. where elements of AM have already been applied to energy planning in order to investigate some of the critical determinants for its successful implementation to date. This assessment then informs a prospective analysis of three West African cases that have ideal characteristics for experimentation with AM to determine to what extent similar concepts have been used, or may be employed in the future. The AM framework also calls for the consideration of local values, which should be open to revision in the face of real situations. To this end, the prospective analysis includes three additional place-sensitive criteria, so as to ensure that the framework remains viable in a different socio-political context. The AM analyses are then extended to include a discussion of learning and innovation in clean energy technologies, drawing from the Chinese, Danish and South African experiences. The results suggest that a strong and consistent political will that is in alignment with an explicit social policy is needed to initiate and implement broad-based electrification plans, but that stakeholder participation is critical to their success. In addition, the adoption of multiple instruments and the selection of a diverse range of energy resources were found to be more effective than an overreliance on a single dominant scheme so as to allow room for policy learning. Furthermore, the results confirm that a holistic approach to managing ecosystems associated with electric power production is a fruitful way to integrate ecological considerations with social and economic factors throughout the development of a project. This type of systemic methodology should also include the building of technological capability and the development of innovation capacity in order to address the unique socio-economic context and the rapidly-changing climatic conditions in West Africa. Finally, the articulation of a planning philosophy that engages the values and sensibilities of the people in a particular place, and that is rooted in them, was found to be a critical factor for increasing the level of public participation in management activities in order to achieve more equitable and democratic outcomes.

L'efficacité énergétique est devenue une métrique clé de la performance des systèmes sur puce numériques, en particulier pour les applications tirant leur énergie de batteries ou de l'environnement. La miniaturisation technologique n'est plus suffisante pour atteindre les niveaux de consommation requis. Ce travail de recherche propose ainsi de nouvelles conceptions de circuits pour la génération d'horloge flexible, la mesure de puissance et de température ainsi que l'intégration de ces blocs au sein de systèmes sur puce complets.Le multiplieur de fréquence innovant en boucle ouverte proposé permet l'adaptation rapide de la fréquence générée (53MHz 0.5V - 889MHz 0.9 V). Sa surface réduite (981µm2) et faible consommation (0.45pJ/cycle à 0.5 V) facilitent son intégration dans des systèmes à basse consommation. Le capteur de puissance instrumente un convertisseur de tension switched-capacitor; validé sur deux architectures différentes, il permet une mesure de la puissance d'entrée et de sortie avec une précision de 2.5% à 6%. Enfin, un nouveau principe de capteur de température est proposé. Il exploite une méthode de calibration par body-biasing sur caisson n et un système numérique intégré pour la compensation de non-linéarité. Enfin, cette thèse illustre la manière dont ces circuits peuvent être intégrés pour assurer la gestion de consommation de systèmes complexes. Un travail de modélisation du body-biasing est proposé, illustrant sa complémentarité avec la gestion de tension d'alimentation. Puis trois exemples de stratégies de gestion de la consommation sont proposées au sein de systèmes complets
Energy efficiency has become a key metric for digital SoC, especially for applications relying on batteries or energy harvesting. Hence, this work proposes new designs for on-chip flexible clock generator, power monitor and temperature sensor as well as the integration of those blocks within complete SoC.The novel open-loop clock multiplier architecture enables fast frequency scaling and is implemented to operate on the same voltage-frequency range as a digital core ((53MHz 0.5V - 889MHz 0.9 V). The achieved extremely low area (981µm2) and power consumption 0.45pJ/cycle 0.5 V) also ease its integration within low power SoC. The proposed power monitor instruments switched capacitor DC-DC converters, which are standard components of low voltage SoCs. The monitor has been demonstrated over two different converters topologies and provides a measurement of both the converter input and output power within 2.5% to 6% accuracy. Last, a new principle of temperature sensor is proposed. It leverages single n well body-biasing for calibration and integrated digital logic for large non-linearity correction. It is expected to achieve within 1C accuracy 0.1nJ / sample and 225 µm2 probe area. Then, this work illustrates how those circuits can be integrated within complex SoCs power management strategies. First, a modeling study of body biasing highlights the benefits it can provide in complement to voltage scaling, accounting for a wide temperature range. Last, three example of power management are proposed at SoC level

The aim of this thesis is to explore the possibility of a dual-purpose electric vehicle. The vehicle should be able to be used for both commuting and racing. It also aims to describe different power limiting methods and their effect on performance. Lastly it hopes to see if the Swedish laws, as written today are reasonable. An electric longboard has been constructed for this purpose and several tests have been performed. A list of goals were set up for the board prototype. These included power output, running time, and that the board should have an audible warning device. The findings demonstrated that all tested power limiting methods worked and that the “Simple power limiting” method provided quickest movement over a fixed distance. Most of the goals were met by the prototype and the board’s two modes worked as planned. While the law is reasonable it can be improved to cover the diversity of electric vehicles. Further work may include better measurements and implementation of a dual microcontroller system.
Syftet med detta arbete är att testa möjligheten med ett två-läges personligt fordon. Fordonenet ska kunna användas för både pendling och tävling. Arbetet har även testat olika energilimeteringsmetoder samt undersökt hur dessa metoder påverkar prestandan. Arbetet har även försökt besvara frågan om det svenska lagarna, som det är skrivna idag, är lämpliga. En elektrisk longboard har konstruerats och flera test utförts. Ett antal målsättningar har definerats för prototypen. Dessa inkluderar: effekt, körtid och att brädan bör ha en ringklocka. Resultaten visar att alla metoder fungerade och att “Simple power limiting” var den som gav snabbast rörelse över den bestämda sträckan. Det flesta av målen nåddes av prototypen och brädans två lägen fungerade som tänkt. Dagens lagar är rimliga men kan förbättras för att täcka mångfalden av elektriska fordon. Framtida arbete kan inkludera bättre mätningar och implementation av ett system med två mikrokontroller.

The aim of this thesis is to explore the possibility of adual-purpose electric vehicle. The vehicle should be ableto be used for both commuting and racing. It also aims todescribe different power limiting methods and their effecton performance. Lastly it hopes to see if the Swedish laws,as written today are reasonable. An electric longboard hasbeen constructed for this purpose and several tests havebeen performed.A list of goals were set up for the board prototype. Theseincluded power output, running time, and that the boardshould have an audible warning device.The findings demonstrated that all tested power limitingmethods worked and that the “Simple power limiting” methodprovided quickest movement over a fixed distance. Most ofthe goals were met by the prototype and the board’s twomodes worked as planned. While the law is reasonable itcan be improved to cover the diversity of electric vehicles.Further work may include better measurements and implementationof a dual microcontroller system.
Syftet med detta arbete är att testa möjligheten med etttvå-läges personligt fordon. Fordonenet ska kunna användasför både pendling och tävling. Arbetet har även testat olikaenergilimeteringsmetoder samt undersökt hur dessa metoderpåverkar prestandan. Arbetet har även försökt besvarafrågan om det svenska lagarna, som det är skrivna idag, ärlämpliga. En elektrisk longboard har konstruerats och fleratest utförts.Ett antal målsättningar har definerats för prototypen. Dessainkluderar: effekt, körtid och att brädan bör ha en ringklocka.Resultaten visar att alla metoder fungerade och att “Simplepower limiting” var den som gav snabbast rörelse över denbestämda sträckan. Det flesta av målen nåddes av prototypenoch brädans två lägen fungerade som tänkt. Dagens lagarär rimliga men kan förbättras för att täcka mångfaldenav elektriska fordon. Framtida arbete kan inkludera bättremätningar och implementation av ett system med två mikrokontroller.

Les limites des ressources d’énergies fossiles et la perspective imminente du changement climatique ont mené les pays de l’Union Européenne à engager une restructuration du secteur électrique vers un approvisionnement en énergie fiable, économique et durable. Dans cette optique de transition énergétique, les systèmes de stockage d’énergie peuvent faciliter l’intégration des énergies renouvelables dans les réseaux électriques. Ils permettent de stocker l’énergie produite par les sources renouvelables pour décaler sa fourniture aux réseaux électriques et compenser les fluctuations aléatoires de la puissance. Par ce lissage de la production des sources intermittentes, les systèmes de stockage transforment ces dernières en centrales mieux contrôlables et plus prévisibles ce qui leur permet de participer aux marchés d’électricité et aux services systèmes.Afin de garantir le respect des plans de production et des engagements pris envers le gestionnaire du réseau, les centrales de production renouvelables équipées d’un système de stockage ont recours à un système de gestion d’énergie. Alors que le contrôle rapproché assure le respect de la consigne instantanée de production, la gestion d’énergie utilise des méthodes d’optimisation sous contraintes issues de la recherche opérationnelle pour planifier le fonctionnement des systèmes de stockage. Le plus souvent, un arbitrage est nécessaire entre les exigences du fonctionnement et la complexité du modèle utilisé. Les modèles de batterie, qui présentent un comportement non-linéaire, doivent être simplifiés en vue de les intégrer dans les algorithmes d’optimisation les plus courants. De plus, les modèles précis et particulièrement ceux qui sont basés sur une modélisation physico-chimique de la batterie exigent des tests de caractérisation chronophages réalisés dans des conditions contrôlées. Finalement, le comportement électrique de la batterie évolue avec son âge ce qui impose un recalage périodique du modèle en fonction du temps.Cette thèse présente une méthodologie d’identification de modèles de batterie en cours de fonctionnement et d’utilisation de ces modèles adaptatifs dans la gestion optimale d’une centrale de production électrique avec stockage. Après un rappel des modèles de batteries, des méthodes d’identification en temps réel issues de la théorie du contrôle sont développées dans le cas d’un modèle des circuits électriques équivalents. L’extraction d’un modèle simplifié pour la gestion d’énergie est décrite et juxtaposée à une analyse de régression directe des données de fonctionnement. Les méthodes d’identification sont testées pour un système de stockage réel de taille industrielle, associé à une centrale photovoltaïque installée sur l’île de La Réunion. L’identification du modèle dans le cadre d’une étude de vieillissement préalablement effectuée au sein du CEA, met en évidence le suivi de l’état de santé de la batterie.En vue de l’intégration des modèles adaptatifs dans la gestion, la formulation des problèmes d’optimisation rencontrés dans la planification d'une centrale photovoltaïque associée à un système de stockage est développée. Des implémentations en programmation linéaire-mixte et en programmation dynamique sont réalisées dans des cas d’études basés sur la participation aux marchés d’électricité ou dans le cadre d’une tarification réglementée, ainsi que la participation aux services système. Afin d’évaluer les performances de ces solutions, une architecture de contrôle pour la centrale est détaillée, et le fonctionnement de la centrale est simulé. Plusieurs configurations du système de gestion sont testées, y compris l’utilisation de modèles fixes ou variables ainsi que la prise en compte ou non du vieillissement de la batterie. Une analyse statistique des résultats obtenus pour différents cas de production photovoltaïque et d’erreurs de prévision montre que l’utilisation des modèles variables présente des avantages
Limited fossil energy resources and the prospect of impending climate change have led the European Union to engage in a restructuring of the electricity sector towards a sustainable, economical and reliable power supply. Energy storage systems have the potential of an enabling technology for the integration of renewable energy sources, which underlies this transition. They allow the delivery of energy produced by a local source to the electric grid to be shifted in time and can compensate random fluctuations in power output. Through such smoothing and levelling, energy storage systems can make the production of variable renewable sources predictable and amenable to control.In order to observe scheduled production and their commitments toward the grid operator, renewable power plants equipped with storage systems make use of an energy management system. While direct control ensures tracking of the current production setpoint, energy management employs constrained optimization methods from operations research to organize the usage of the storage systems. The complexity of the storage system model used in optimization must frequently be adapted to the specific application. Batteries show non-linear state-dependent behavior. Their model must be simplified for use in the most common optimization algorithms. Moreover, precise battery models based on physical modelling require time-consuming controlled testing for parameterization. Lastly, the electrical behavior of a battery evolves with aging which calls for regular recalibration of the model.This thesis presents a methodology for on-line battery model identification and the use of such adaptive models in optimal management of an electrical plant with energy storage. After a summary of battery models, observer methods for on-line identification based on control theory are developed for the case of an equivalent circuit model. The extraction of a simplified model for energy management is described and compared to direct regression analysis of the operational data. The identification methods are tested for a real industrial-sized storage system operated in a photovoltaic power plant on the island of La Réunion. Model identification applied to data from an earlier battery aging study shows the use of the method for tracking the state-of-health.The formulation of optimization problems encountered in the production scheduling of a photovoltaic power plant with energy storage is developed incorporating the adaptive battery models. Mixed-integer linear programming and dynamic programming implementations are used in case studies based on market integration of the plant or regulated feed-in tariffs. A simulation model based on the outline of the plant control architecture is used to simulate the operation and evaluate the solutions. Different configurations of the management system are tested, including static and variable battery models and the integration of battery aging. A statistical analysis of the results obtained for multiple cases of photovoltaic production and forecast error shows the advantage of using variable battery models in the study case

Au-delà du nœud technologique CMOS BULK 28nm, certaines limites ont été atteintes dans l'amélioration des performances en raison notamment d'une consommation énergétique devenant trop importante. C'est une des raisons pour lesquelles de nouvelles technologies ont été développées, notamment celles basées sur Silicium sur Isolant (SOI). Par ailleurs, la généralisation des architectures complexes de type multi-cœurs, accentue le problème de gestion de la consommation à grain-fin. Les technologies CMOS FD-SOI offrent de nouvelles opportunités pour la gestion de la consommation en permettant d'ajuster, outre les paramètres usuels que sont la tension d'alimentation et la fréquence d'horloge, la tension de body bias. C'est dans ce contexte que ce travail étudie les nouvelles possibilités offertes et explore des solutions innovantes de gestion dynamique de la tension d'alimentation, fréquence d'horloge et tension de body bias afin d'optimiser la consommation énergétique des systèmes sur puce. L'ensemble des paramètres tensions/fréquence permettent une multitude de points de fonctionnement, qui doivent satisfaire des contraintes de fonctionnalité et de performance. Ce travail s'intéresse donc dans un premier temps à une problématique de conception, en proposant une méthode d'optimisation du placement de ces points de fonctionnement. Une solution analytique permettant de maximiser le gain en consommation apporté par l'utilisation de plusieurs points de fonctionnement est proposée. La deuxième contribution importante de cette thèse concerne la gestion dynamique de la tension d'alimentation, de la fréquence et de la tension de body bias, permettant d'optimiser l'efficacité énergétique en se basant sur le concept de convexité. La validation expérimentale des méthodes proposées s'appuie sur des échantillons de circuits réels, et montre des gains en consommation moyens allant jusqu'à 35%
Beyond 28nm CMOS BULK technology node, some limits have been reached in terms of performance improvements. This is mainly due to the increasing power consumption. This is one of the reasons why new technologies have been developed, including those based on Silicon-On-Insulator (SOI). Moreover, the standardization of complex architectures such as multi-core architectures emphasizes the problem of power management at fine-grain. FD-SOI technologies offer new power management opportunities by adjusting, in addition to the usual parameters such as supply voltage and clock frequency, the body bias voltage. In this context, this work explores new opportunities and searches novel solutions for dynamically manage supply voltage, clock frequency and body bias voltage in order to optimize the power consumption of System on Chip.Adjusting supply voltage, frequency and body bias parameters allows multiple operating points, which must satisfy the constraints of functionality and performance. This work focuses initially at design time, proposing a method to optimize the placement of these operating points. An analytical solution to maximize power savings achieved through the use of several operating points is provided. The second important contribution of this work is a method based on convexity concept to dynamically manage the supply voltage, the frequency and the body bias voltage so as to optimize the energy efficiency. The experimental results based on real circuits show average power savings reaching 35%

The environmental management of watersheds presents a complex governance issue due to their large spatial scales that include overlapping jurisdictions, competing interests in resource use, and lack of coordination among stakeholders. The Rideau Canal, spanning 200 km between the cities of Ottawa and Kingston, is an interesting case study as it is a multi-watershed system over which municipal, provincial and federal governments have authority. However, these governments have been unsuccessful in addressing system-wide issues such as shoreline development, erosion and invasive species that have significantly impacted the ecological integrity of the canal. A shift toward polycentric governance, which are systems of multi-scale governance, in which well-informed publics can contribute to the Rideau Canal’s management is required. This thesis examines how co-governance can be conceptualized for the RC by (1) analyzing convergences in stakeholder perspectives about the environment and governance, and (2) comparing collaborative causal mapping exercises with various stakeholders to current government engagement efforts. A tiered co-governance framework that intentionally links existing small-scale activities to system-wide formal venues of knowledge sharing could democratize environmental governance on the Rideau Canal to improve its management. Beyond its practical contributions, this research also contributes to developing the academic literature on co-governance for multi-watershed waterways that have both constructed and natural aspects.

Les développements récents dans le domaine des circuits intégrés (IC) à basse tension ont ouvert la voie à des dispositifs électroniques économes en énergie dans un réseau mondial en plein essor appelé l’internet des objets (IoT) ou l’internet des choses (IoE). Cependant, la durabilité de tous ces capteurs interconnectés est compromise par le besoin constant d’une batterie embarquée – qui doit être rechargée ou remplacée – ou d’un récupérateur d’énergie à rendement très limité. La consommation d’énergie des systèmes électroniques grand public actuels est en effet cinquante fois plus élevée que celle d’un collecteur d’une taille de l’ordre du cm 2 , ou limitée à quelques mois sur une petite batterie. Cela contraint la viabilité de solutions fonctionnant à l’échelle d’une vie humaine. Les systèmes sur puce (SoCs) à venir nécessitent donc de relever le défi de cette lacune énergétique en optimisant l’architecture, de la technologie au niveau du système. L’approche technique de ce travail vise à démontrer la faisabilité d’un SoC efficient, ultra-basse tension (ULV) et ultra-basse puissance (ULP) utilisant exclusivement les dernières directives industrielles en matière de technologies FD-SOI (Fully Depleted Silicon On Insulator) 28 nm et 22 nm. Plusieurs SoCs multi-domaines basés sur des cœurs ARM sont implémentés pour démontrer des stratégies de réveil basées sur les entrées des capteurs. Ainsi, en optimisant l’architecture du système, en sélectionnant et en concevant correctement les composants avec des caractéristiques technologiques choisies de manière adéquate, et en ajustant soigneusement l’implémentation physique, on obtient un SoC entièrement optimisé en énergie
Recent developments in the field of low voltage integrated circuits (IC) have paved the way towards energy efficient electronic devices in a booming global network called the internet-of-things (IoT) or the internet-of-everything (IoE). However, the sustainability of all these inter- connected sensors is still undermined by the constant need for either an on-board battery – that must be recharged or replaced – or an energy harvester with very limited power efficiency. The power consumption of present consumer electronic systems is fifty times higher than the energy available by cm 2-size harvester or limited to a few months on a small battery, thus hardly viable for lifetime solutions. Upcoming systems-on-chip (SoCs) must overcome the challenge of this energy gap by architecture optimizations from technology to system level. The technical approach of this work aims to demonstrate the feasibility of an efficient ultra-low-voltage (ULV) and ultra-low-power (ULP) SoC using exclusively latest industrial guidelines in 28 nm and 22 nm fully depleted silicon on insulator (FD-SOI) technologies. Several multi-power-domain SoCs based on ARM cores are implemented to demonstrate wake up strategies based on sensors inputs. By optimizing the system architecture, properly selecting and designing compo- nents with technology features chosen adequately, carefully tuning the implementation, a fully energy-optimized SoC is realized

El impacto acelerado de las actividades humanas está causando el aumento de los daños a los sistemas de soporte vital de la Tierra. En consecuencia, la necesidad de un cambio hacia el uso sostenible de los recursos naturales y los ecosistemas se ha defendido como una necesidad urgente por científicos y tomadores de decisiones. Esta tesis trata sobre las condiciones institucionales necesarias para dicho cambio en sistemas socio-ecológicos, a través de un estudio de caso en profundidad: la región de Doñana. En particular, la tesis se centra en la necesidad de transiciones desde estrategias de mando y control hacia enfoques más flexibles y adaptativos para la elaboración de políticas y la toma de decisiones, en particular, la gobernanza adaptativa y la gestión adaptativa. Para ello, se abordan tres preguntas de interés de investigación, amplias e interrelacionadas, utilizando un marco teórico que combina elementos de la dependencia de la trayectoria institucional y la teoría de la resiliencia. La primera pregunta de investigación se centra en la evaluación de la utilidad de un programa de investigación-acción cuyo objetivo fue introducir principios de gestión adaptativa en la interfaz investigación-gestión de la región de Doñana. La segunda pregunta se centra en mejorar la comprensión de las raíces de la rigidez institucional en sistemas socio-ecológicos maladaptativos. La tercera pregunta se centra en el potencial explicativo del emprendimiento y los discursos en su relación con los intereses político-económicos y el poder, como factores que contribuyen a la formación de sistemas socio-ecológicos a nivel local. El enriquecimiento del conocimiento adquirido durante el programa de investigación-acción con el análisis en profundidad de las limitaciones institucionales arraigadas en factores históricos, permitieron la identificación de una serie de posibles vías que pueden ayudar a la transición hacia la gobernanza y la gestión adaptativas en la región de Doñana. Asimismo, este enriquecimiento permitió una especulación informada sobre el papel potencial de programas de investigación-acción como el que se describe en esta tesis, para cumplir con (y complementar) los requisitos para la participación pública y el aprendizaje social de la legislación de la Unión Europea – en particular, la Directiva Marco del Agua.

The complexity of distributed computing systems has raised two unprecedented challenges for system management. First, various customers need to be assured by meeting their required service-level agreements such as response time and throughput. Second, system power consumption must be controlled in order to avoid system failures caused by power capacity overload or system overheating due to increasingly high server density. However, most existing work, unfortunately, either relies on open-loop estimations based on off-line profiled system models, or evolves in a more ad hoc fashion, which requires exhaustive iterations of tuning and testing, or oversimplifies the problem by ignoring the coupling between different system characteristics (ie, response time and throughput, power consumption of different servers). As a result, the majority of previous work lacks rigorous guarantees on the performance and power consumption for computing systems, and may result in degraded overall system performance. In this thesis, we extensively study adaptive performance/power management and power-efficient performance management for distributed computing systems such as information dissemination systems, power grid management systems, and data centers, by proposing Multiple-Input-Multiple-Output (MIMO) control and hierarchical designs based on feedback control theory. For adaptive performance management, we design an integrated solution that controls both the average response time and CPU utilization in information dissemination systems to achieve bounded response time for high-priority information and maximized system throughput in an example information dissemination system. In addition, we design a hierarchical control solution to guarantee the deadlines of real-time tasks in power grid computing by grouping them based on their characteristics, respectively. For adaptive power management, we design MIMO optimal control solutions for power control at the cluster and server level and a hierarchical solution for large-scale data centers. Our MIMO control design can capture the coupling among different system characteristics, while our hierarchical design can coordinate controllers at different levels. For power-efficient performance management, we discuss a two-layer coordinated management solution for virtualized data centers. Experimental results in both physical testbeds and simulations demonstrate that all the solutions outperform state-of-the-art management schemes by significantly improving overall system performance.

This thesis discusses a power electronics module (PEM) that is used to extract power from a human energy harvesting generator according to the user’s desired input power, and stores all of the extracted energy into an appropriately sized battery while staying within the charging limitations of the battery. The PEM can temporarily store the peak power produced by the generator allowing the reduction in the size of the battery required to the average power production level of the generator. The battery’s safety and longevity is maintained by charging them at the constant current and constant voltage rate. The design of the two-stage PEM, the requirements of the Energy Storage Capacitor (ESC) and battery size are discussed. The two controllers that control the PEM are explained and the different operating modes of the controllers are reviewed. A two-stage prototype digitally controlled average current mode control Boost converter and average current mode controlled Buck converter were designed and experimental waveforms were captured to test and validate the control theories used in the PEM. A Voltage Adaptive Gain compensator was used to optimize the closed loop response of both the Boost and Buck converters over their respective output and input voltage ranges. The DC efficiency of the prototype was measured with the peak efficiency of the Boost converter equal to 93% and the peak efficiency of the Buck converter measured at 93.7%. The total PEM system efficiency was measured at 87.9% at an input power level of 10 watts. The AC efficiency of the PEM was also measured with a peak efficiency of 91% with Vin = 15 V at Rin = 60 Ω. The software considerations for an embedded system, including power consumption and timing of real time events are reviewed. A software flow chart and timing diagram are provided to help visualize the sequence of the code. A design chart for selection of the size and voltage rating of the ESC was created. An experimental comparison of a single stage design without energy storage capability and the current PEM design was performed, with a power limited source, in order to show the effectiveness of the PEM and controllers at maximizing the power extraction from the generator. The PEM design was able to extract 50% more power than the single stage converter without energy storage capability.
Thesis (Master, Electrical & Computer Engineering) -- Queen's University, 2012-03-20 01:25:20.986

博士
國立交通大學
電子研究所
99
Adaptive power management designs are presented in this thesis including an all digital controlled linear regulator and an adaptive power control technique. Each one is essentially a stand-alone attachment for digital integrated circuit blocks while they can also be adopted jointly to have more flexibility on power control. Since the analog regulator design has suffered a lot from technology advancing, an all digital controlled variable output linear voltage regulator that supports voltage scaling is presented in this thesis. A test chip had been fabricated on UMC 65nm standard CMOS technology. The developed digital voltage regulator has a 99.8% current efficiency with only 164.5_A quiescent current. The area of the control system is about 300um^2. A response time constraint has been developed as well to provide a design guideline for (all) the digital control system. It describes the correlation between required speed of the digital control system, the output performance, and the size of the decoupling capacitor. A proposed time interleaving control can have trade-off between these parameters. The adaptive power control technique can utilize unused slack and reduce power. The switching state determination mechanism is the core technique replacing the critical path replica to detect circuit speed. It is intrinsically tolerant of PVT variations. The circuit speed can be altered by dynamically configuring the size of the power gating devices and hence reduce power. A test chip had been fabricated on UMC 90nm standard CMOS technology. The area and power overhead are both around 1% relative to a 32-bit multiplier. The proposed technique can achieve averages of 56.5% slack utilization, 12.39% net power reduction, and 87.5% leakage reduction. The adaptive power management designs are discussed on 2D planar and TSV 3DIC applications with temperature-aware power management methodology. A multi-layer power delivery structure is presented as well when going to TSV 3DIC applications.

abstract: Nearly 60% of the world population uses a mobile phone, which is typically powered by a system-on-chip (SoC). While the mobile platform capabilities range widely, responsiveness, long battery life and reliability are common design concerns that are crucial to remain competitive. Consequently, state-of-the-art mobile platforms have become highly heterogeneous by combining a powerful SoC with numerous other resources, including display, memory, power management IC, battery and wireless modems. Furthermore, the SoC itself is a heterogeneous resource that integrates many processing elements, such as CPU cores, GPU, video, image, and audio processors. Therefore, CPU cores do not dominate the platform power consumption under many application scenarios. Competitive performance requires higher operating frequency, and leads to larger power consumption. In turn, power consumption increases the junction and skin temperatures, which have adverse effects on the device reliability and user experience. As a result, allocating the power budget among the major platform resources and temperature control have become fundamental consideration for mobile platforms. Dynamic thermal and power management algorithms address this problem by putting a subset of the processing elements or shared resources to sleep states, or throttling their frequencies. However, an adhoc approach could easily cripple the performance, if it slows down the performance-critical processing element. Furthermore, mobile platforms run a wide range of applications with time varying workload characteristics, unlike early generations, which supported only limited functionality. As a result, there is a need for adaptive power and performance management approaches that consider the platform as a whole, rather than focusing on a subset. Towards this need, our specific contributions include (a) a framework to dynamically select the Pareto-optimal frequency and active cores for the heterogeneous CPUs, such as ARM big.Little architecture, (b) a dynamic power budgeting approach for allocating optimal power consumption to the CPU and GPU using performance sensitivity models for each PE, (c) an adaptive GPU frame time sensitivity prediction model to aid power management algorithms, and (d) an online learning algorithm that constructs adaptive run-time models for non-stationary workloads.
Dissertation/Thesis
Doctoral Dissertation Electrical Engineering 2018

碩士
國立臺北科技大學
車輛工程系所
100
The increase of oil price has become a global long-term trend. Major automotive manufactures started to put more resource in the research and development of new energy vehicles. Electric vehicle (EV) has the advantage of zero missions. However, its cruising distance is often limited due the insufficient battery energy density. Range extended electric vehicle (REEV) can extend its cruising distance by using the range extender which consists of engine and generator, i.e. genset. Thus, range extender is considered as a short term alternative solution to solve the current problem of insufficient cruising distance of EV. For a given driving pattern, equivalent fuel consumption minimization will be used to design the proposed adaptive power management control strategy, such that fuel consumption can be effectively reduced for different driving patterns. Driver only needs to provide the approximate estimation of the traveling distance to plan the reference SOC trajectory for discharging the battery. Self-organizing fuzzy controller will be used to adaptively adjust the equivalence factor which is used to convert the electric power usage to equivalent fuel consumption. A cost function of instantaneous fuel consumption is minimized to obtain the optimum power split between the genset and battery. Simulation results show that the proposed algorithm can improve the fuel economy and reduce the average charging/discharging power of the battery.

碩士
國立臺灣大學
電子工程學研究所
100
The rectifier is one of the most important circuits in power harvesting systems. In this thesis, three works about the 2.4-GHz rectifier have been done. In the first research, an optimization method of the conventional single-type rectifier is proposed. Through this optimization process, two rectifiers with the peak PCE equal to 51.43% and 18.13% are designed. In second work, a high-efficiency adaptive rectifier is proposed. By this structure, the input power range is expanded which means the limitation of the wireless operating distance can be relaxed a lot. According to [6], the optimal PCE region is the RF input power range where the PCE is larger than 80% of the peak PCE. Therefore, the optimal PCE region of the proposed rectifier is 11dB which is improved by over 3dB compared with other works. In final work, a wireless charging circuit which consists of a high efficiency rectifier and a power management circuit is proposed. This charging circuit can provide stable 1.8V voltage and 3mA current to back-end circuits.

Energy-related costs are among the major contributors to the total cost of ownership of data centers and high-performance computing (HPC) clusters. As a result, future data centers must be energy-efficient to meet the continuously increasing computational demand. Constraining the power consumption of the servers is a widely used approach for managing energy costs and complying with power delivery limitations. In tandem, virtualization has become a common practice, as virtualization reduces hardware and power requirements by enabling consolidation of multiple applications on to a smaller set of physical resources. However, administration and management of data center resources have become more complex due to the growing number of virtualized servers installed in data centers. Therefore, designing autonomous and adaptive energy efficiency approaches is crucial to achieve sustainable and cost-efficient operation in data centers. Many modern data centers running enterprise workloads successfully implement energy efficiency approaches today. However, the nature of multi-threaded applications, which are becoming more common in all computing domains, brings additional design and management challenges. Tackling these challenges requires a deeper understanding of the interactions between the applications and the underlying hardware nodes. Although cluster-level management techniques bring significant benefits, node-level techniques provide more visibility into application characteristics, which can then be used to further improve the overall energy efficiency of the data centers. This thesis proposes adaptive runtime power and resource management techniques on multi-core systems. It demonstrates that taking the multi-threaded workload characteristics into account during management significantly improves the energy efficiency of the server nodes, which are the basic building blocks of data centers. The key distinguishing features of this work are as follows: We implement the proposed runtime techniques on state-of-the-art commodity multi-core servers and show that their energy efficiency can be significantly improved by (1) taking multi-threaded application specific characteristics into account while making resource allocation decisions, (2) accurately tracking dynamically changing power constraints by using low-overhead application-aware runtime techniques, and (3) coordinating dynamic adaptive decisions at various layers of the computing stack, specifically at system and application levels. Our results show that efficient resource distribution under power constraints yields energy savings of up to 24% compared to existing approaches, along with the ability to meet power constraints 98% of the time for a diverse set of multi-threaded applications.

碩士
國立臺北科技大學
電機工程系所
94
With the fast development of SOC, the processor and peripheral device controller are able to be integrated into one chip. The systems with such an SOC chip and sophisticated operating system are called embedded system and are frequently made portable. Since the power of a portable device is typically supplied by a battery system, the designers of embedded system have paid effort to the techniques on lengthening operation time of a portable device. Dynamic power management system is one of those techniques. We implement an embedded system with learning-tree-prediction-based dynamic power management. The consumed power of the embedded system is obtained by way of a self-made current measuring module. The measured power consumption are sent to a PC through RS-232. The experimental results show that after applying the learning tree prediction, the power consumption of the embedded system is effectively reduced. Also, after modifying the learning tree in our own way, the power consumption is further reduced.

碩士
國立中山大學
電機工程學系研究所
98
In a multi-hop mobile ad hoc network (MANET), IEEE 802.11 power management may fail if power-saving (PS for short) stations are out of synchronization. To fix this problem, [8, 12, 15, 16] proposed various cyclic quorum-based power management (CQPM) protocols, which, however, may also completely fail if some PS stations have different schedule repetition intervals (SRIs). This implies that, from the viewpoint of duty cycle, CQPM is non-adaptive. Hence the authors of [1] proposed the first adaptive power management protocol, called AQEC. However, in AQEC, the values of SRI must be squares. This implies the adaptiveness (i.e., the number of tunable SRIs) of AQEC is very limited. Hence the authors of [14] proposed the first fully adaptive power management protocol, called HQS. However, in HQS, the duty cycle of a station is not optimal. To conquer all these problems, we propose the OFAA (optimal fully adaptive asynchronous) power management protocol, which has the following attractive features. (i) By means of novel beacon interval structures and the factor-hereditary quorum space techniques, OFAA ensures that two PS neighbors can discover each other in finite time regardless of their clock difference and their individual SRIs. (ii) Given the maximum SRI, Smax, the number of tunable SRIs for every PS station is Smax. (iii) The idle duty cycles for all SRIs are minimal. (iv) The time complexity of OFAA neighbor maintenance is constant. (v) A cross-layer SRI adjustment scheme is proposed such that a PS station can adaptively tune its SRI according to traffic QoS requirements. Both theoretical analysis and simulation results show that OFAA achieves better energy efficiency than existing adaptive CQPM protocols, including AQEC [1] and HQS [14].

碩士
國立中山大學
資訊工程學系研究所
97
In the past few years, due to the rapid advance in technology and the aid of 3D graphics applications the world of 3D graphics is rapidly expanding from desktop computers and dedicated gaming consoled to handheld devices, such as cellular phones, PDAs, laptops etc.,. However, unlike traditional desktop computers and gaming consoles, mobile computing devices typically have slower processors that have less capability for handling large computation-intensive workloads like 3D graphics application. In addition, the power consumption is one of the major design specifications to realize the 3D graphics accelerating engine for mobile devices because handheld batteries have limited lifetimes. Moreover, the size of chip is depend on the Moore’s Law: The number of transistors in a chip are double in every eighteen months. Even though the produce cost is decrease, but the capacity of battery cannot increase like the transistors. Therefore, how to reduce power consumption by using efficient power management techniques has become a very important research topic in 3D graphics SoC design. For 3D graphics applications, dynamic voltage and frequency scaling (DVFS) is a good candidate to reduce the power consumption of 3D graphics accelerating engine. So many relative papers have researched in how to accurately predict the workload and scale the voltage and frequency. The prediction policy can divide into History-based predictor [1] and Frame-structure predictor [2-4]. The History-based predictor predicts the latter frame workload by previous frame workload to scale the voltage, and the frame-structure predictor performs offline and then determine the different kind of frame for an application. A table is used to save the mapping of different kind of frame to the voltage, and then the voltage is scaled according to the mapping table. A lot of researchers put the power management policy in software i.e. processors, but our proposed workload prediction scheme has been realized into the hardware circuit. Therefore, it can not only reduce the overhead of processor but also quickly adjust the voltage and frequency of 3D graphics accelerating engine. Our prediction policy is one of the History-based predictor ,and it is an adaptive PID predictor [5-6] in which the parameters of Proportional controller and Integral controller can be adaptively adjusted so that it can obtain more accurate prediction results than non-adaptive predictor. In general, the workload that the selected voltage can handle is usually over than the predicted workload. That is, actual workload is usually less than predicted workload. So that the slack time will be generated. We can utilize the slack time through Inter-frame compensation [7-10] to save more energy while maintaining the similar output quality. We use a simple policy to adaptively select the parameters for compensation between the frames to simplify the hardware architecture of the power management policy. Experimental results show that, we can get more energy saving and more accurate workload prediction when the adaptive PI predictor and adaptive Inter-frame compensation are utilized.

碩士
國立臺灣科技大學
電機工程系
104
Energy efficiency is a critical issue for battery-driven mobile devices. With the increasing needs of graphic mobile games on such devices, an on-line power governor for both CPUs and GPUs is highly motivated. This study proposes an adaptive on-line CPU-GPU governor of games on mobile devices to reduce the energy. In contrast to existing governors, an adaptive frequency scaling framework is presented to switch the performance-driven or energy-driven to minimize the energy without user attention. Also, to adjust the dynamic of quality during game stages, an on-line learning is applied to our framework to predict the required quality and to sense the relations between quality and CPU/GPU frequencies. The idea is implemented on Google Nexus 7 and evaluated with real-world gaming apps. The results show that we can save up to 20% system energy (included the network, screen, and systemidle power) for a high frame rate game and maintaining a stable user experience.

碩士
國立中山大學
資訊工程學系研究所
98
As time goes by rapid development of 3D graphics technique and 3C portable product output, 3D graphics have been widely applied to handheld devices, such as notebooks, PDAs, and smart cellular phones. Generally, to process 3D graphics applications in mobile devices, processor needs strong capability of handling large computational-intensive workloads. Complex computation consumes a great quantity of electric power. But the lifetime of handheld device battery is limited. Therefore, the cost, to satisfy this demand, will be shortening the supply time of device battery. Moreover, Moore’ law said that the number of transistors in a chip is double in every eighteen months. But these days the advance in manufacturing batteries still cannot get up with the advance in developing processors. In addition, the improvement of chip size has led to more small, supply voltage of kernel processor in portable device. Considering system efficiency and battery lifetime simultaneously increase the difficulty of designing power management scheme. So, how to manage power effectively has become one of the important key for designing handheld products. For 3D graphics system, dynamic voltage and frequency scaling (DVFS) is one of good solutions to implement power management policy. DVFS needs an efficient online prediction method to predict the workload of frames and then appropriately adjust voltage and frequency for saving energy consumption. Consequently, a lot of related papers have proposed different prediction policy to predict the executing workload of 3D graphics system. For instance, the existing prediction policies include signature-based[1], history-based[3] and proportion-integral-derivative (PID)[14] methods, but most of designers put power management in software, i.e. processors. This solution not only slows power management to get the information about executing time of graphic processing unit (GPU), but also increases the operating overhead of CPU in handheld system. In this paper, we propose a power management workload prediction scheme with a framework of using proportion-integral (PI) controller to be a master controller and fuzzy controller to be a slave controller, and then implement it into hardware circuit. Taking advantage of fuzzy conception in fuzzy controller is to adjust the proportional parameter in PI controller, the shortage of traditional PI controller that demands on complicated try-and-error method to look for a good proportional and integral parameters can be avoided so that the adaption and forecasting accuracy can be improved. Besides, Uniform Window-size Predictor 1 (UW1) is also implemented as an assistant manner. Using UW1 predictor appropriately can improve the prediction trend to catch up with the trend of real workload. Experimental results show that our predictor improves prediction accuracy about 3.8% on average and saves about 0.02% more energy compared with PI predictor[18]. Circuit area and power consumption only increases 6.8% percent and 1.4% compared with PI predictor. Besides, we also apply our predictor to the 3D first person game, Quake II, in the market. The result shows that our predictor is indeed an effective prediction policy. The adaption can put up with the intense workload variation of real game and adjust voltage and frequency precisely to decrease power consumption and meet the purpose of energy saving.

博士
國立臺灣大學
電子工程學研究所
105
As a result of the continuing growth of miniaturization and integration technologies in the semiconductor industry and the emergence of the internet of things (IoT), the continuing trend is to reduce the power supply voltage to below 1 V and increase the power supply current greatly exceed the 1000 mA per chip. Therefore, the next-generation power management integrated circuits must be able to deal with the challenge of the ultra-low voltage and fast changing load profiles. Since the traditional topology of the current-mode pulse width modulation (PWM) has its bandwidth and stability limitation, the transient response and the power quality cannot be further improved. Therefore, this dissertation is aiming at the high speed adaptive controlling techniques for current-mode PWM converters. In this dissertation, two novel high speed adaptive controlling techniques are proposed. 　　The “synchronous double-pumping (SDP) technique” can dramatically improve the transient response of the conventional current-mode PWM converters. The prototype chip was fabricated using TSMC 0.35 μm CMOS process. The experimental measured recovery time is 2.4 μs and 2.6 μs, respectively, in response to the 400mA step-up and step-down load changes. Those are improved by a factor of 8.33 and 8.23, respectively. 　　The “dynamically adjusted triple-feedback-loop controlling (DATFLC)” technique used an extra dynamic voltage adjusting feedback loop, which is comprising of a variable gain peak current sampler (VGPCS), adaptive load transient accelerator (ALTA) and a dynamic analog voltage elevator (DAVE), to accelerate the transient response of the conventional current-mode PWM converters. Besides, this technique used a transient accelerated zero-current detector (TAZCD) to further accelerate the load step-down response by means of allowing the reversed inductor current to discharge the converter output through the lower gate of the power stage at the beginning periods of the load step-down response. Moreover, the physical implementation of the proposed DATFLC technique is really compact, and the chip area overhead is only 7.41% to the core area. The prototype chip was fabricated using TSMC 0.25 μm HV CMOS process. The experimental measured recovery time is 3.2 μs and 3.0 μs, respectively, in response to the 400mA step-up and step-down load changes. Those are improved by a factor of 7.38 and 8.20, respectively.

Freshwater scarcity is a dire problem for exposed human societies and natural ecosystems—a problem expected to grow worse with anticipated climate change. Reverse osmosis (RO) desalination is currently the most energy-efficient and ubiquitous desalination process used for freshwater production in water-scarce regions. The synergy of high solar radiation and significantly reduced costs in photovoltaics (PV) creates the opportunity for PV to become a dominant and sustainable solution for powering the energy-intensive process of desalination and reducing greenhouse gas emissions.While photovoltaic-powered reverse osmosis (PVRO) is a promising technological solution, several significant challenges must be further addressed to sustain high RO performance. First, the inherently intermittent nature of solar energy generation can adversely affect the freshwater conversion process and thereby decrease water recovery and quality. Furthermore, global desalination capacity is dominated by large-scale plants, whereas PVRO systems are currently limited to small-scale systems. Thus, to truly integrate renewable energy with desalination systems in an impactful way, there is a need to explore pathways for modifying the RO process to enable flexible operation on a large-scale, in response to power variability. Furthermore, the techno-economic feasibility of flexible, renewable-powered RO processes and the potential benefits that could be provided to variable renewable energy (VRE) plants and the electric grid warrants investigation. Brine minimization is another major challenge for sustainable desalination. Brine management is especially an issue for inland desalination plants. Novel approaches that are less costly and less energy intensive are needed to facilitate minimal and zero liquid discharge. To enable high-salinity desalination, several variations of osmotically assisted RO, which each surpass the pressure limitation of conventional RO, have been proposed in the literature but require further assessment. The promise of these enhanced RO approaches entails a reduction in energy consumption when compared with thermal desalination methods. The primary deliverables and novel contributions of this thesis include the development of (i) design, simulation, and cost optimization models for variable-powered, variable-salinity RO systems, (ii) module-scale, cost-optimization models for enhanced RO technologies that reduce transmembrane osmotic pressure to enable high-salinity desalination and brine minimization, (iii) examining the effects of cyclic reverse osmosis on inorganic scaling mitigation, and (iv) quantifying the availability of unconventional, alternative water sources to alleviate local water scarcity in the contiguous US. First, the techno-economic feasibility of PV-powered RO desalination plants in the Gulf region was assessed using Hybrid Optimization Model for Electric Renewables (HOMER) and Desalination Economic Evaluation Program (DEEP) to model both the power system and desalination system, respectively. Subsequently, an hourly simulation model for desalination was developed to replace the use of DEEP in the workflow. Grid-connected and off-grid cases with combinations of PV, batteries, and diesel generators were evaluated primarily by the levelized cost of electricity (LCOE) and levelized cost of water (LCOW). The shortcoming of conventional and PV-powered RO is that variable power compromises cumulative water production, which in turn increases water costs. Thus, we proposed the concept of active-salinity-control reverse osmosis (ASCRO) which enables control of the transmembrane osmotic pressure and water production in response to variable power. The ASCRO system dynamically controls energy consumption by operating across a range of feed salinity, allowing it to shift over a wide range of pump feed flows and pressures. To accomplish this, ASCRO utilizes feedwater from both low- and high-salinity sources. Enabling a dynamic power consumption profile can enhance demand-response capabilities, compensating for stressors on the grid. Moreover, ASCRO can improve the integration of renewable energy (RE) by responding to power fluctuations without compromising permeate production. This system can include on-site RE and energy storage to power the ASCRO plant and provide services to the grid. We considered the following grid-connected scenarios: 1) ASCRO, 2) ASCRO and battery storage, 3) ASCRO and photovoltaics (PV), and 4) ASCRO, battery storage, and PV. The LCOW was minimized by providing load-shifting and regulation capacity services in the California Independent System Operator (CAISO) market. We quantified that the ASCRO plant can ramp from minimum to maximum load within 84 seconds, which is adequate for participation in fast-timescale markets. The LCOW for these scenarios ranged from 49 – 59 cents/m³. We also present sensitivity analyses showing the effects of capital cost, CAISO market prices, and PV size on LCOW. To investigate alternative pathways to minimal and zero liquid discharge, low-salt rejection reverse osmosis (LSRRO), cascading osmotically mediated reverse osmosis (COMRO), and osmotically assisted reverse osmosis (OARO) were comparatively assessed via module-scale, cost optimization models to gain an accurate perspective of the performance differences between each of these configurations. We quantified the optimal LCOW of each technology for the case of desalinating feedwater at 70 g/L at 75% recovery, which would result in a brine concentration near 250 g/L, a level that allows further treatment with crystallizers. For baseline scenarios, LCOW results for OARO, COMRO, and LSRRO were 5.14, 7.90, and 6.63 $/m³ of product water, respectively, while the corresponding specific energy consumption (SEC) values were 10.31, 12.77, and 28.90 kWh/m³. A sensitivity analysis is also presented. Additionally, we sought to examine the possibility of whether adaptive RO operation could provide the added benefit of fouling mitigation. Using the Pitzer model, nucleation theory, and dissolution kinetics to guide a set of bench-scale fouling experiments, CaSO₄-NaCl solution, supersaturated with respect to gypsum, was fed through a membrane test cell to determine nucleation induction times, rates of flux decline, and scale reversal. Lastly, a geospatial analysis was conducted to estimate volumes of water deficits and potential alternative water sources for the contiguous US. Namely, wastewater effluent, brackish groundwater, agricultural drainage water, and produced water were considered in this analysis as alternatives for alleviating water scarcity. We formulated a conservative estimate of groundwater availability based on environmental flow limits. Additionally, agricultural drainage volumes were estimated based on USGS water use data. Overall, the results showed that water deficits amounted to an equivalent daily capacity of 149 million m³/day—nearly 50% more than the desalination capacity of the world in 2020. Furthermore, the total availability of alternative water sources was estimated to be between 192 – 240 million m³/day, but most of this volume was not in the same location as deficits. Thus, 58 – 65% of national water deficits would have to be alleviated via long-range transport. Additionally, the potential for integrating desalination and water reuse by interconnecting existing RO plants with wastewater treatments plants was also assessed.

The demand to integrate more features has significantly increased the complexity and power consumption of smart portable devices. Therefore extending the battery life-time has become a major challenge and new approaches are required to decrease the power consumed from the source. Traditionally the focus has been on reducing the dynamic power consumption of the digital circuits used in these devices. However as process technologies scale, reducing the dynamic power has become less effective due to the increased impact of the leakage power. Alternatively, a more effective approach to minimize the power consumption is to continuously optimize the ratio of the dynamic and leakage power while delivering the required performance. This works presents a novel power-aware system for dynamic minimum power point tracking of digital loads in portable applications. The system integrates a dc-dc converter power-stage and the supplied digital circuit. The integrated dc-dc converter IC utilizes a mixed-signal current program mode (CPM) controller to regulate the supply voltage of the digital load IC. This embedded converter inherently measures the power consumption of the load in real-time, eliminating the need for additional power sensing circuitry. Based on the information available in the CPM controller, a minimum power point tracking (MiPPT) controller sets the supply and threshold voltages for the digital load to minimize its power consumption while maintaining a target frequency. The 10MHz mixed-signal CPM controlled dc-dc converter and the digital load are fabricated in 0.13µm IBM technology. Experimental results verify that the introduced system results in up to 30% lower power consumption from the battery source.