With the accessible data reaching zettabyte level, CMOS technology is reaching its limit for the data hungry applications. Moore's law has been reaching its depletion in recent studies. On the other hand, von Neumann architecture is approaching the bottleneck due to the data movement between the computing and memory units. With data movement and power budgets becoming the limiting factors of today's computing systems, in-memory computing using emerging non-volatile resistive devices has attracted an increasing amount of attention. A non-volatile resistive device may be realized using memristor, resistive random access memory (ReRAM), phase change memory (PCM), or spin-transfer torque magnetic random access memory (STT-MRAM). Resistive devices integrated into crossbar arrays simultaneously supports both dense storage and energy-efficient analog computation, which is highly desirable for processing of big data using both low-power mobile devices and high-performance computing (HPC) systems. However, analog computation is vulnerable and may suffer from robustness issues due to variations such as, array parasitics, device defects, non-ideal device characteristics, and various sources of errors. These non-ideal factors directly impact the computational accuracy of the in-memory computation and thereby the application level functional correctness. This dissertation is focused on improving the robustness and reliability of analog in-memory computing. Three directions are mainly explored: data layout organization techniques, software and hardware co-design, and hardware redundancy. Data layout organization aims to improve the robustness by masking the data to hardware according to the behavior of defective devices. Software and hardware co-design mitigates the impact by modifying the data in the neural networks or image compression applications to become amenable to device defects and data layout organizations. Hardware redundancy utilized multiple resistive device to realize each data, so each device can be programmed with different value and realize the data accurately with lower overhead.

Energy storage systems such as batteries and supercapacitors store electrical energy in the form of chemical energy and release it when required. Among the various electrode materials, manganese oxides (MnOx) are promising electrode materials for these devices. Despite its outstanding theoretical capacitance, Mn-based oxide electrodes have several limitations that impede their electrochemical performance. Understanding how the charges are efficiently stored in the electrodes or across the electrode/electrolyte interface is crucial for developing advanced electrode material in the field of energy storage applications. The goal of my thesis is to develop and apply synchrotron-based scanning transmission X-ray microscopy (STXM) to investigate changes in the oxidation state of Mn and their spatial distributions in MnOx electrodes in the context of energy storage and release. To achieve high- precision qualitative and quantitative STXM identification and mapping of different MnOx species, calibrated and high-quality reference Mn 2p and O 1s NEXAFS (near edge X-ray absorption fine structure) spectra were measured. In collaboration with Wenjuan Yang and her PhD supervisor, Prof. Igor Zhitomirsky, I performed ex-situ STXM studies on Mn3O4-based supercapacitor electrode materials to investigate the influence of different synthesis methods and activation protocols on the charging behavior and capacitance performance. In collaboration with Pablo Ingino and his supervisor, Prof. Martin Obst (Bayreuth University), and my colleague, Dr. Chunyang Zhang, I helped develop a three-electrode, microfluidic-based flow electrochemical device for in-situ STXM. This device was used to electrodeposit MnO2 on the working electrode (WE) and track the oxidation state and morphological changes by STXM while scanning the potential of the cell in different electrolyte pH. The in-situ STXM studies showed a spontaneous reduction of the initially deposited MnO2 resulting from the local pH change at the WE. Additionally, a significant change from a quasi-uniform MnO2 film to a dendritic MnO2 structure was observed at oxidative potential. This dendritic growth resulted from dissolution/redeposition of MnO2 during charging/discharging processes, indicating a partial reversibility of dissoluble Mn species. The ex-situ and in-situ STXM studies I performed provide mechanistic insights that will help further improve Mn oxides-based electrodes and their applications as energy storage devices. / Thesis / Doctor of Philosophy (PhD)

No description available.

missing pages: 69, 76, 83, 99-100, 114, some pages have faint ink due to the condition of the original

The growing performance gap between CPUs and sustainable disk I/O is a major hurdle in supporting modern applications. As the CPUs become faster, this gap is projected to worsen, thus making it a critical problem that should be addressed with high priority. Although efficient algorithms have alleviated this problem, the mechanical nature of the disk places physical limits on the achievable speedup. On the other hand, newer technologies such as flash memory promise significant improvements in access time, power consumption, and storage density. However, the mature disk technology offers the most favorable cost per bit ratio. Since replacing standard hard disks with flash disks is prohibitively expensive, hybrid hard disks augment the standard hard disk with a small amount of flash memory. By exploiting the beneficial aspects of both technologies they aim to provide breakthrough increase in performance. Nevertheless, hybrid hard disks pose several significant design challenges. Effective and efficient algorithms to manage the flash, the disk, and interaction between them are required. To facilitate rapid and easy exploration of the design space for hybrid hard disk algorithms we present the design and implementation of a flexible and extensible simulator that models hybrid hard disks. The simulator is flexible in that it models several configurations in which the flash and the magnetic medium interact. The simulator is extensible in that it provides a simple framework to plug in several algorithms to manage the hybrid hard disk. We validate our simulator and analyze the performance of the hybrid hard disk for real workloads. / Master of Science

Global efforts to reduce greenhouse gas emissions, particularly CO2, have led countries to focus on decarbonizing the transportation sector. Moving towards electric vehicles (EVs) is necessary to reduce emissions, however despite EV technological advancements they have shortcomings in both performance and longevity. Supercapacitors are similar to batteries, however their ability to easily charge and discharge at much higher rates makes them excellent devices to work in tandem with batteries to advance their collective performance capabilities in EVs. Traditional metal-based supercapacitor materials remain to be high cost, non-renewable, and often environmentally toxic. On the other hand, quinones are organic materials considered as promising candidates for organic electrodes due to the redox activity, low cost, ease of structural modifications, nontoxicity, and renewability. To overcome quinone challenges with low electrical conductivity and dissolution in electrolyte, polymerizing quinones has become a popular modification. Conducting polymers (CPs) are increasing in interest as their -conjugated structures provide efficient electron transfer and good electrical conductivity. In the work of this master’s thesis, two types of materials were developed for supercapacitor applications; a polyimide made from alternating units of the quinones 3,4,9,10-perylenetetracarboxylic dianhydride (PTCDA) and 2,6-diaminoanthraquinone (DAAQ) known as poly-perylene-3,4,9,10-tetracarboxydiimide-anthraquinone (PPA), and a truncated analogue of PPA comprised of PTCDA and two molecules of 2-diaminoanthraquinone (2-AAQ), termed N,N′-bis(2-anthra-quinone)]-perylene-3,4,9,10- tetracarboxydiimide (PDI-DAQ). All the original redox-active sites were retained following a facile synthesis to achieve fast multi-electron transfer mechanisms. These materials both were used to prepare composite electrodes with a low-cost carbon black (Ketjenblack) via simple and scalable preparation methods. Capacitances reached up to 377 F/g at 5 mV s-1 with a capacitance retention of 63.9% after 10,000 cycles at 100 mV/s. This work demonstrates the impressive energy storage capabilities of novel organic molecules in supercapacitors with low-cost carbon black to improve the performance of next-generation EVs. / Thesis / Master of Applied Science (MASc) / Worldwide efforts to reduce greenhouse gas emissions, like CO2, have made the world focusing on making more environmentally sustainable transportation methods, such as switching to electric vehicles (EVs). However, EVs still face performance and longevity issues due to the limitations of the batteries used. Batteries are not designed to charge and discharge quickly, however supercapacitors, which are like batteries, can charge and discharge much faster, making them a great match to incorporate into EVs alongside batteries. Traditional metal supercapacitor materials are costly and non-sustainable, but organic molecules like quinones offer a much cheaper, sustainable solution. Modifying quinones along with the addition of cheap carbon additives can vastly improve its energy storage performance and long-term usage. With future scalability in mind, this work demonstrates the potential for organic materials to potentially be used to enhance the performance of next generation EVs.

The thesis, Öppet Lager, (Open Storage) attaches a public structure to Moderna Museet’s off-site storage to bring together the showing and storing of art. As it is inevitable that large institutions cannot show their full collections, many of their works of art will be stored away from the public. Moderna Museet currently has 283 artworks on display at Skeppsholmen which is around 0.25 percent out of their collection of 140 000 pieces. As collections continue to grow the percentage of what is shown decreases, and the storage spaces fill up. To open the storages of art Öppet Lager addresses the conflict between showing and storing art and how objects could be displayed without compromising the administrative functions of transportation, conservation, and research. By proposing an alternative way to store art, the project aims to create greater access to works of art that would otherwise be sleeping deep in storages.

Laboratory scale activated sludge experiments were conducted on primary effluent municipal wastewater to evaluate the effects of anoxic selectors on controlling activated sludge bulking and foaming. These experiments were conducted with two pilot plants; a three stage anoxic selector preceding a complete mix system (experimental unit) and a complete mix system (control unit). Successful selector operation requires balancing two conflicting requirements; obtain a high substrate concentration in the selector while achieving a high substrate removal efficiency in the selector. The high substrate concentration enables rapid substrate uptake to occur predominately by floc forming microorganisms while the high substrate removal efficiency ensures that a feed-starve cycle is created whereby filamentous microorganisms are selected against. The reported metabolic mechanisms responsible for substrate uptake in the selector are the formation of internal storage products and high rate metabolism. As presented by Jenkins et al., (1993) small amounts of substrate are oxidized in the selector during the formation of internal storage products. Hence, large quantities of substrate can be removed while reducing only small amounts of the terminal electron acceptor. The internal stores are metabolized in the main biological reactor only after the exogenous substrate has been exhausted. High rate metabolism in the selector results in larger amounts of substrate oxidation. Consequently, for successful selector operation large quantities of the terminal electron acceptor must be reduced. The anoxic selector pilot unit successfully reduced activated sludge settleability and biological foams relative to the control unit. Results from this study indicate that the mode of substrate removal was influenced by the initial selector floc load. This is in general agreement with the findings by Goel and Gaudy (1968) and Gaudy and Gaudy (1988) on oxidative assimilation in activated sludge treatment. The floc load depicts the instantaneous organic loading in the selector irrespective of hydraulic retention time. Results from this study further indicate that for lower floc loadings substrate storage is predominate. Alternatively, at higher floc loadings high rate substrate metabolism is predominate. Therefore, it is hypothesized that for selector zones with high enough F/M ratios to promote rapid substrate uptake, the mechanism predominately responsible for substrate removal is influenced by the floc loading. / Master of Science

The purpose of this investigation is to find which season of the year organic matter increases most in the bottom deposits of Lake Dallas, the reason for the increase, and the amount of organic matter increase from year to year. It is hoped that this study will be beneficial in understanding the conditions in artificial reservoirs.

I takt med en ökande befolkning ökar användningen av energi. Samtidigt som energianvändandet ökar, avvecklas kärnkraftverken och därmed ökar kolkraftverkens användning vilket leder till utsläpp av främst koldioxid. Många industrier släpper ut mängder av överskottsvärme i naturen utan att den återanvänds. Ett sätt att ta tillvara på överskottsvärme, som annars går till spillo, är att lagra den. Om värme kan lagras och användas vid en annan tidpunkt kan den ersätta andra energikällor och onödiga utsläpp kan förhindras. Det finns idag tre olika metoder att lagra värmeenergi. Dessa är sensibelt värme, latent värme och kemisk värme. Inom varje metod finns olika system som beskrivs vidare i denna rapport. Bharat Forge Kilsta Kilsta är ett smidesföretag i Karlskoga. Deras smidesugn avger stora mängder värme som dels går till lokaluppvärmning men en del av värmen går till spillo. Skulle överskottsvärmen, som nu går till spillo, kunna lagras på ett effektivt sätt skulle både miljömässiga och kostnadsmässiga besparingar kunna göras. Syftet med rapporten är att redogöra och jämföra olika värmelagringsmetoder i en litteraturstudie för att se vilken typ som passar för industriell överskottsvärme i fallet med Bharat Forge Kilsta. Målet är att översiktligt redovisa olika lagringsmetoder samt olika system inom dessa med avseende på lagringskapacitet och kostnad. Utifrån simulering och reglering av bergrumslager och ackumulatortankar kan en passande metod, med avseende på energidistribution och energieffektivitet samt kostnad, för det specifika fallet väljas. Den mest utvecklade och kommersiellt använda metoden är sensibelt värme, den latenta och kemiska värmelagringen är fortfarande i forskning- och utvecklingsstadiet då de är mer kostsamma. Val av lagringsmetod avgörs utifrån lagringskapacitet, lagringstemperatur, kostnad, geografisk placering samt lagringslängd. Sensibelt värme passar bäst till långtidslagring, vid lägre temperaturer och där lagringskapaciteten måste vara stor till ett lågt pris. Latent och kemisk värme passar bäst för högre temperaturer då värmeförlusterna är små och energidensiteten är hög, kostnaden för dessa är dock hög och de tillämpas enbart i liten skala än så länge. Ur litteraturstudien kunde vissa system uteslutas, de system som skulle passa en industri som Bharat Forge Kilsta var bergrum och ackumulatortank. Resultatet visade att bergrummen har störst värmeförluster jämfört med den totala energin, däremot är lagringskapaciteten större. För att garanterat tillgodose värmebehovet vid extremdagar är det mest lämpligt att använda bergrummen. Kostnadsmässigt är de befintliga tankarna bäst lämpade, däremot klarar de enbart tillgodose värmebehovet i sex timmar vid extrembelastning. Om de befintliga tankarna används som system och 200 m3 tanken tilläggsisoleras kan omkring 100 000 kr per år sparas, räknat med att förlusterna skulle ersätta inköpt fjärrvärme och att skillnaden i värmeförluster enbart sker vinterhalvåret. Återbetalningstiden var kortast för de befintliga takarna, 1,4 år medan en ny ackumulatortank hade längst återbetalningstid, 3,2 år. / When the population increases also the energy use will rise. At the same time the nuclear power plants is decommissioned and the use of coal-fired power plants increases, which leads to large amount of mainly carbon dioxide emissions. Many industries get a lot of excess heat that is released in the nature instead of being reused. One way to reuse excess heat could be to store the heat in a suitable storage for later use. If the excess heat can be stored and be used at a different time it can replace other energy sources and decrease the emissions. Today there is three ways to storage heat, they are sensible heat, latent heat, and chemical heat. In each method there are different systems, these will be described further in this report. Bharat Fore is a large forging company in Karlskoga, Sweden. From their furnace a lot of heat is emitted, some of the heat is used to heat the buildings, but still a lot of excess heat goes to waste.  The aim of this report is to compare different heat storage systems and see which one is best suited to industrial excess heat. The goal is to investigate if there is any heat storage method that is effective and cost-saving that fits a larger industry. The purpose of this work is to do a literature study to account and compare different heat storage methods to find the best suitable system for the case with Bharat Forge Kilsta. The goal is to present different storage methods and the different system for each method with respect of cost and storage capacity. From simulation and regulation find the best fitting method for the real case with respect of cost, efficient and storage capacity. The most developed and commercially used method is the sensible heat. Latent heat and chemicals are very costly and still in the research and development stage. Geographic location, using area and operating temperature is parameters that need to be considered when choosing heat storage system. Sensible heat is best suited for long-term storage, at lower temperatures and when the storage capacity needs to be large to a small cost. Latent and chemical heat is best suited for higher temperatures because the heat losses are small and the energy density is high and they are only applied in small scale for now. The result of the literature study showed that storage tanks and cavern storage is most fitting for the case with Bharat Forge Kilsta. The cavern has much larger heat loss compared to the total energy, however the storage capacity is much larger. To guarantee that the heat requirements when there are extreme days it is most appropriate to use the cavern as heat storage. From a coast view it is most fitting to use the already existing tanks, however they could only cater the heat requirement for six hours of heat peak when the production is not running. If the existing tanks is used as heat storage, and the 200 m3 tank will be additional insulated, if the heat loss, in the winter, is replaced with purchased district heating as much as 100 000 SEK per year could be spared. The payback time is shortest for the existing tanks, 1.4 years and almost 3.2 years for the new storage tank.