Thesis (M.Sc.Eng.) PLEASE NOTE: Boston University Libraries did not receive an Authorization To Manage form for this thesis or dissertation. It is therefore not openly accessible, though it may be available by request. If you are the author or principal advisor of this work and would like to request open access for it, please contact us at open-help@bu.edu. Thank you. / The spectra of low piano tones have long been known to display unusual, inharmonic characteristics. Surprisingly, precisely this property allowed for fairly successful generation of artificial piano tones by the frequency modulation (FM) synthesis techniques pioneered at Stanford in the 1970s. Though FM synthesis techniques are largely empirical in nature, the close correspondence between the spectra of the FM simulation of piano tones and their actual spectra suggests the hypothesis that there may be a deeper physical connection between frequency modulation and the timbre of string instruments. Recent, highly sophisticated numerical simulations of the physics of pianos have reproduced these characteristic spectra using a geometrically exact form of the equations for vibrating strings. We take these equations as a starting point for an analytical investigation into the connection between modulated waves and instrumental timbre. The content of the investigation is primarily theoretical, establishing approaches and equations which may be used for further exploration. We proceed by deriving the equations of motion from variational principles. The Lagrangian is given careful treatment so that the transverse and longitudinal directions of motion are clearly separated and their coupling is collected in a single interaction term. Approximate equations of motion are then derived. These are shown to have a structure that supports modulated wave behavior. A general form of the solution is offered which has the form a traveling wave plus a correction term. Understanding the correction term demands the solution of a complex set of nonlinear integral equations, though suggestions are made on how determining certain properties of the integral equations would be sufficient for significant insight into the role of modulation effects in the production of distinct musical timbres. Finally, another form of the solution is offered, one more amenable to iterative solution techniques but less transparent in terms of frequency-domain behavior. A discussion regarding the initial conditions and the requirements for the convergence of either of these forms of the solution is given. / 2031-01-01

Artificial photosynthesis utilises solar-light for clean fuel H2 production and is emerging as a potential solution for renewable energy generation. Photocatalytic systems that combine a light harvester and catalysts in one-pot reactor are promising strategies towards this direction. Yet, most of the reported systems function by consuming excess amount of expensive sacrificial reagents, preventing commercial development. In this thesis, carbon nitrides (CNx) have been selected as non-toxic, stable and low-cost photocatalysts. CNx are first introduced as efficient light harvesters, to couple alcohol oxidation with proton reduction, in the presence of a Ni-based molecular catalyst. This system operated in a single compartment while the oxidation and reduction products were collected in the solution and gaseous phases, respectively, demonstrating a closed redox system. In the presence of an organic substrate and absence of a proton reduction catalyst, photoexcited CNx was found to accumulate long-lived "trapped-electrons", which enables decoupling oxidation and reduction reactions temporarily and spatially. This allows solar H2 generation in the dark, following light exposure, replication light and dark cycle of natural photosynthesis in an artificial set-up. The stability of the designed system was found to be limited by the Ni-based molecular catalyst, and the spectroscopic studies revealed electron transfer from CNx to catalyst as the kinetic bottleneck. Graphene based conductive scaffolds were introduced to the CNx-Ni system, to accelerate the rate of electron transfer from CNx to the Ni catalyst. Time-resolved spectroscopic techniques revealed that introducing these conductive binders enabled better electronic communication between CNx and Ni, resulting in significantly enhanced photocatalytic activity. To improve the solar-light utilisation and the photocatalytic performance of bulk CNx, a straightforward ultra-sonication approach was introduced. This pre-treatment was found to break aggregates of bulk CNx, and the resulting activated CNx had significantly improved activity. The activated CNx showed record activities per gram of the material used, for H2 evolution with a molecular Ni catalyst. The use of abundant waste sources instead of organic substrates was investigated in the presence of activated CNx. The system demonstrated to photoreform purified and raw lignocellulose samples into H2 in the presence of various H2 evolution catalysts over a wide range of pH.

The evolution of disease resistance in plants occurs within a framework of interacting phenotypes, balancing natural selection for life-history traits along a continuum of fast-growing and poorly defended, or slow-growing and well-defended lifestyles. Plant populations connected by gene flow are physiologically limited to evolving along a single axis of the spectrum of the growth-defense trade-off, and strong local selection can purge phenotypic variance from a population or species, making it difficult to detect variation linked to the trade-off. Hybridization between two species that have evolved different growth-defense trade-off optima can reveal trade-offs hidden in either species by introducing phenotypic and genetic variance. Here, I investigated the phenotypic and genetic basis for variation of disease resistance in a set of naturally formed hybrid poplars. The focal species of this dissertation were the balsam poplar (Populus balsamifera), black balsam poplar (P. trichocarpa), narrowleaf cottonwood (P. angustifolia), and eastern cottonwood (P. deltoides). Vegetative cuttings of samples were collected from natural populations and clonally replicated in a common garden. Ecophysiology and stomata traits, and the severity of poplar leaf rust disease (Melampsora medusae) were collected. To overcome the methodological bottleneck of manually phenotyping stomata density for thousands of cuticle micrographs, I developed a publicly available tool to automatically identify and count stomata. To identify stomata, a deep con- volutional neural network was trained on over 4,000 cuticle images of over 700 plant species. The neural network had an accuracy of 94.2% when applied to new cuticle images and phenotyped hundreds of micrographs in a matter of minutes. To understand how disease severity, stomata, and ecophysiology traits changed as a result of hybridization, statistical models were fit that included the expected proportion of the genome from either parental species in a hybrid. These models in- dicated that the ratio of stomata on the upper surface of the leaf to the total number of stomata was strongly linked to disease, was highly heritable, and wass sensitive to hybridization. I further investigated the genomic basis of stomata-linked disease variation by performing an association genetic analysis that explicitly incorporated admixture. Positive selection in genes involved in guard cell regulation, immune sys- tem negative regulation, detoxification, lipid biosynthesis, and cell wall homeostasis were identified. Together, my dissertation incorporated advances in image-based phenotyping with evolutionary theory, directed at understanding how disease frequency changes when hybridization alters the genomes of a population.

Objectives: To evaluate the influence of five different displays on the diagnosis of artificial caries-like lesions on restored and unrestored tooth surfaces. Materials & Methods: 60 extracted human teeth (30 premolars and 30 molars) were selected. All molars had class II cavities prepared and restored. Half of the premolars and molars were randomly selected and a 7mm² area was exposed to a demineralizing solution for 120 days. Phantoms with four teeth (two premolars and two molars) were created. Periapical radiographs were obtained on RVG 6100 digital sensor (Kodak Dental Systems, New York, USA). The images were evaluated under subdued lighting on five viewing displays (Barco MDNC-3321, NEC Wide, Apple iPad Pro, Microsoft Surface Pro 4 and Dell flat panel monitor) by three observers using a five-point rating scale. Sensitivity (Sn), specificity (Sp), accuracy (Ac) and Receiver Operating Characteristic (ROC) curves and their areas under the curves (AUC) were calculated and compared by Analysis of variance and post-hoc Tukey test. Intraobserver and interobserver agreement were accomplished after a three month interval observational.. The observer agreement was evaluated with the Cohen’s kappa test. Results: Even though the tablets slightly improved accuracy over the medical monitors, there was no statistically significant difference in sensitivity, specificity, accuracy or AUC among the five monitors when the same tooth group was considered (p< 0.05) with the exception of the iPad Pro which had a specificity p-value of 0.014 when comparing the two materials within the iPad Pro. Conclusion: There is no perceivable disadvantage to utilizing a higher resolution tablet viewing platform for plane images. No advantage to the medical grade monitors over the tablets or the consumer monitor. Both tablets improved detection accuracy on the unrestored premolars. However further evaluation especially in a clinical lighting setting is warranted to evaluate a displays impact on observer performance within common clinical practice lighting parameters.

[FeFe]-Hydrogenases (H2ases) are metalloenzymes that can catalyze the reversible reduction of protons to molecular hydrogen as part of the metabolism of certain cyanobacteria and green algae. Due to the low availability of the enzyme, synthetic complexes that mimic the natural active site in structure, function and activity are highly sought after. In this thesis, a number of [FeFe]-H2ases active site model complexes were synthesized to answer open questions of the active site and to develop unprecedented bio-inspired proton reduction catalysts. The first part describes the synthesis and the protonation properties of a [Fe2(μ-adt)(CO)4(PMe3)2] (adt = azadithiolate) complex which contains two basic sites that are similar to those found in the enzyme active site. Unusual kinetic factors give rise to four discrete protonation states. The twofold protonated state is the first model complex that simultaneously carries a proton at the azadithiolate nitrogen and a bridging hydride at the Fe-Fe bond. In the second part, a model complex with an unprecedented amine ligand was synthesized and studied. In analogy to the enzyme active site, the labile amine ligand is expelled after electrochemical reduction. The third part describes a series of model complexes with electronically different aromatic dithiolate ligands. It is demonstrated in one case that the tuning of the ligand by electron-withdrawing substituents results in proton reduction catalysis at an overpotential that is lower than that required by the non-substituted parent compound. The design and the synthetic work towards a new ruthenium-diiron dyad for light-driven hydrogen production are presented in the fourth part. In the final part, differently isotope-labelled mixed valent Fe(I)-Fe(II) model complexes were synthesized, in particular the unprecedented 15N labelled analogue, with the aim to provide EPR-spectroscopic references that will allow the elucidation of the nature of the central atom in the dithiolate bridge of the [FeFe] hydrogenase active site.

Context. Reinforcement Learning (RL) is a time consuming effort that requires a lot of computational power as well. There are mainly two approaches to improving RL efficiency, the theoretical mathematics and algorithmic approach or the practical implementation approach. In this study, the approaches are combined in an attempt to reduce time consumption.\newline Objectives. We investigate whether modern hardware and software, GPGPU, combined with state-of-art Evolution Strategies, CMA-Neuro-ES, can potentially increase the efficiency of solving RL problems.\newline Methods. In order to do this, both an implementational as well as an experimental research method is used. The implementational research mainly involves developing and setting up an experimental framework in which to measure efficiency through benchmarking. In this framework, the GPGPU/ES solution is later developed. Using this framework, experiments are conducted on a conventional sequential solution as well as our own parallel GPGPU solution.\newline Results. The results indicate that utilizing GPGPU and state-of-art ES when attempting to solve RL problems can be more efficient in terms of time consumption in comparison to a conventional and sequential CPU approach.\newline Conclusions. We conclude that our proposed solution requires additional work and research but that it shows promise already in this initial study. As the study is focused on primarily generating benchmark performance data from the experiments, the study lacks data on RL efficiency and thus motivation for using our approach. However we do conclude that the GPGPU approach suggested does allow less time consuming RL problem solving.

This is a report about machine learning in the field of computer science. The problem handled is prediction of energy consumption in district heating systems. Prediction of energy consumption in district heating systems is a delicate problem because of the social behaviours, weather and distribution time that has to be accounted for. One algorithm is introduced and three different experiments are made to determine if the algorithm is useful. The results from the experiments were good. This report differs in approach to the problem then other reports found in this field. The difference is that this report tries to handle social behaviours and looks at a decentralized view of the problem instead of centralized. / Denna rapport är om maskininlärning och hur mna kan använda en maskinlärningsalgoritm för att förutspå konsumption i fjärrvärmenät. Rapporten skiljer sig markant i synsätt jämt emot andra rapporter i ämnet genom att den tittar även på de sociala faktorerna.

Thesis is focused on safety and reliability of space products. It describes assessment methods and applies them on small commercial satellite.

Aus der Einführung: "Stehen wir am Rande einer bio-nanotechnologischen getriebenen Revolution, die unsere Art zu leben, zu arbeiten und miteinander umzugehen grundlegend verändern wird? Welchem gesellschaftspolitischen, wirtschaftlichen und technologischen Wandel haben wir uns zu stellen? Langfristige Entwicklungszyklen (Kondratieff, Schumpeter) führen zur nachhaltigen Weiterentwicklung der Zivilisation. Mittelfristige Entwicklungen wie die Trends Globalisierung, Urbanisierung, Digitalisierung (Miniaturisierung) und Humanisierung (Individualisierung), die immer stärker unser Umfeld und Handeln beeinflussen führen zu ganzheitlichen, weltumspannenden Grundtendenzen der gesellschaftlichen Weiterentwicklung. Die technologischen "Enabler" Computing, Biotechnology, Artifical Intelligence, Robotik, Nanotechnology, Additive Manufacturing und Design Thinking wirken beschleunigend auf die gesellschaftlichen Entwicklungen ein. Die technologischen Möglichkeiten beschleunigen sowohl gesellschaftspolitische Zyklen und zivilisatorische Anpassungen. Durch rasanten technologischen, wissenschaftlichen Fortschritt, zunehmende Globalisierungswirkungen, beschleunigte Urbanisierung und aber auch politischer Interferenzen sind die Veränderungsparameter eines dynamischen Geschäftsumfelds immer schnellere Transformationen ausgesetzt. Alle diese Richtungen zeigen das unsere gesellschaftliche Entwicklung inzwischen stark durch die Technik getrieben ist. Ob dies auch heißt, dass wir den Punkt der Singularität (Kurzweil) absehbar erreichen ist dennoch noch offen. ..."

Legitimacy is about meeting society's expectations and as we live in a society that is constantly changing, the perception of what is legitimate or not is also changing. The main purpose of an auditor in today's society is to review financial reports and guarantee their quality. In recent years, technological developments have changed the profession of auditors and previous research shows that digitalization has made it easier for the auditors as certain parts of the audit process have been automated. The purpose of the study was to examine how AI in audit firms will affect the legitimacy of auditors. The focus has been on the audit companies' working process and how the process is going to change with the help of AI, and finally how the auditors believe this will affect the legitimacy of the audit companies. Qualitative interviews have been used on auditors, to be able to get into the depth and obtain personal thoughts from the respondents. The results of the study show that the introduction of AI will have a positive impact on the audit process since it is assumed that standardized and time-consuming tasks will be automated. Furthermore, this means that auditors will be able to perform more accurate analyzes and get more time over to the customers, which means that the audit companies are assumed to be able to increase their legitimacy. Therefore, the conclusion is that AI will help audit companies with their legitimacy in the future.