The Multi-Layer Perceptron (MLP) is one of the most widely applied and researched Artificial Neural Network model. MLP networks are normally applied to performing supervised learning tasks, which involve iterative training methods to adjust the connection weights within the network. This is commonly formulated as a multivariate non-linear optimization problem over a very high-dimensional space of possible weight configurations. Analogous to the field of mathematical optimization, training an MLP is often described as the search of an error surface for a weight vector which gives the smallest possible error value. Although this presents a useful notion of the training process, there are many problems associated with using the error surface to understand the behaviour of learning algorithms and the properties of MLP mappings themselves. Because of the high-dimensionality of the system, many existing methods of analysis are not well-suited to this problem. Visualizing and describing the error surface are also nontrivial and problematic. These problems are specific to complex systems such as neural networks, which contain large numbers of adjustable parameters, and the investigation of such systems in this way is largely a developing area of research. In this thesis, the concept of the error surface is explored using three related methods. Firstly, Principal Component Analysis (PCA) is proposed as a method for visualizing the learning trajectory followed by an algorithm on the error surface. It is found that PCA provides an effective method for performing such a visualization, as well as providing an indication of the significance of individual weights to the training process. Secondly, sampling methods are used to explore the error surface and to measure certain properties of the error surface, providing the necessary data for an intuitive description of the error surface. A number of practical MLP error surfaces are found to contain a high degree of ultrametric structure, in common with other known configuration spaces of complex systems. Thirdly, a class of global optimization algorithms is also developed, which is focused on the construction and evolution of a model of the error surface (or search spa ce) as an integral part of the optimization process. The relationships between this algorithm class, the Population-Based Incremental Learning algorithm, evolutionary algorithms and cooperative search are discussed. The work provides important practical techniques for exploration of the error surfaces of MLP networks. These techniques can be used to examine the dynamics of different training algorithms, the complexity of MLP mappings and an intuitive description of the nature of the error surface. The configuration spaces of other complex systems are also amenable to many of these techniques. Finally, the algorithmic framework provides a powerful paradigm for visualization of the optimization process and the development of parallel coupled optimization algorithms which apply knowledge of the error surface to solving the optimization problem.

In this work, chemical bonding models are developed and extended by the aid of the quantum-chemical position-space analysis. The chemical bonding models are then utilized to rationalize and predict the structure and conducting properties of MgAgAs-type and other intermetallic compounds. Additionally, new position-space bonding indicators are developed.

Made available in DSpace on 2014-06-11T19:22:18Z (GMT). No. of bitstreams: 0 Previous issue date: 2005-12-06Bitstream added on 2014-06-13T19:48:20Z : No. of bitstreams: 1 napoleao_prm_me_rcla.pdf: 1035994 bytes, checksum: 37c800df7ae7f2456ecd217e1c2b5f1d (MD5) / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES) / Este estudo teve como objetivo analisar a distribuição espacial e as relações entre o Índice de Condições de Vida (ICV) e o Índice de Criminalidade Urbana (ICU) nos municípios da Região Administrativa de Campinas (SP) no ano de 2000. A partir do emprego de métodos e técnicas de classificação de dados socioeconômicos foi possível correlacioná-los em um Sistema de Informações Geográficas (SIG). A análise espacial permitiu estabelecer as possíveis relações entre o ICV e o ICU na região e verificar se os municípios com maior ICV foram os que apresentaram o maior índice de criminalidade urbana, em função do crescimento das desigualdades sociais no período. / The objective of this work was to analyze the regional distribution and the relations between Life Level Rates (ICV) and Urban Criminal Rates (ICU) in Campinas Administrative Area (SP), in the year of 2000, using classifying techniques for socioeconomic data in a Geographical Information System (GIS). The spatial analysis evidenced some connections between the ICV and the ICU in that area. It has been verified that cities with larger ICV are the ones with higher urban criminal rates, due to the increase of social inequality during the period.

The main subject of the thesis is a marketing communication strategy in the automotive industry. First part of the thesis defines theory of marketing communication strategy and instruments of the communication mix. The practical part describes Mercedes-Benz Czech republic Inc. and propose marketing communications based on marketing situational analysis and analysis of an external and internal environment.

The ultimate objective of this research was to develop a low cost, reliable system that would lead to early detection of lung cancer. Tests involved the quantitation of gaseous metabolic emissions from immortalized lung cancer cell lines in order to correlate the chemical markers to be of cancerous origin. The specific aims of the project were the study of gas emissions in selected cancer cell lines and identification of volatile organic compounds (VOCs) in them. Disadvantages of earlier studies were that the measurements were not real time or state specific so that molecular identification was often inconclusive. Furthermore the methods of study used in the past were not quantitative, which limited their practicality for medical applications. We felt the need to prove or disprove these earlier results using a new technique. The method we proposed is different and unique when compared to previous methods because cell lines have not been studied extensively for cancer markers. We have studied cancer cell lines which are adherent, immortalized cultures originating from primary tumors obtained from patients with no prior treatment for lung cancer. We have used an alternative method for the spectrometric analysis and quantitation of the selected chemical markers. The pre-concentration method involved a Purge and Trap unit with a thermal desorber where the vapor concentration was enhanced. The concentrated head space gases were analyzed using a Gas Chromatograph ?? Mass Spectrometer setup. This setup eliminated the bulky apparatus used in earlier studies. It is simpler in design and more comprehensive so that external factors such as patient??s diet, habitat and lifestyle do not contribute to our study of recognition of cancer markers. Based on the results obtained in the above experiments, a more comprehensive, inexpensive study of lung cancer related markers could be made. The first section, after giving an introduction to lung cancer, goes on to explain the background work done by other researchers on cancer. The third section gives a detailed explanation of the experimental setup. This is followed by all the tests conducted with corresponding results. The final section deals with the conclusions drawn from all experiments.

To understand the processes affecting the abundance of wild populations is a fundamental goal of ecology and a prerequisite for the management of living resources. Variable abundance, however, makes the investigation of ecological processes challenging. Recruitment, the process whereby new individuals enter a given stage of a ?sh population, is a highly variable entity. I have confronted this issue by developing methodologies speci?cally designed to account for, and ecologically interpret, patterns of variability in recruitment. To provide the necessary context, Chapter 2 begins with a review of the history of recruitment science. I focus on the major achievements as well as present limitations, particularly regarding environmental drivers. Approaches that include explicit environmental information are contrasted with time-varying parameter techniques. In Chapter 3, I ask what patterns of variability in pre-recruit survival can tell us about the strength of density-dependent mortality. I provide methods to investigate the presence of density-dependent mortality where this has previously been hindered by highly variable data. Stochastic density-independent variability is found to be attenuated via density dependence. Sources of recruitment variability are further partitioned in Chapter 4. Using time-varying parameter techniques, signi?cant temporal variation in the annual reproductive rate is found to have occurred in many Atlantic cod populations. Multivariate state space models suggest that populations in close proximity typically have a shared response to environmental change whereas marked differences occur across latitude. Hypotheses that could result in consistent changes in productivity of cod populations are tested in Chapter 5. I focus on a meta-analytical investigation of potential interactions between Atlantic cod and small pelagic species, testing aspects of the cultivation-depensation hypothesis. The ?ndings suggest that predation or competition by herring and mackerel on egg and larval cod could delay recovery of depleted cod populations. Chapter 6 concludes with a critical re?ection on: the suitability of the theories employed, the underlying assumptions of the empirical approaches, and the quality of the data used in my thesis. Application of ecological insights to ?sheries management is critically evaluated. I then propose future work on recruitment processes based on methods presented herein.

Orientador: Marcos César Ferreira / Banca: Sergio dos Anjos Ferreira Pinto / Banca: Magda Adelaide Lombardo Fruehauf / Banca: Diana Sarita Hamburger / Banca: Teresa Gallotti Florenzano / Este estudo teve como propósito principal aplicar e avaliar metodologias para análise fractal da expansão de espaços urbanos metropolitanos, tendo como caso o aglomerado da região metropolitana de São Paulo, Brasil. O estudo foi realizado a partir de mapas temáticos e imagens orbitais relativos ao período 1905-2001, aos quais foram aplicados três métodos para a estimativa da dimensão fractal: densidade de preenchimento (Dd), perímetro-área (DPA) e perímetro-resolução (DPR). A partir da análise fractal dos mapas e imagens, foram avaliados a dinâmica temporal do preenchimento dos vazios urbanos e da expansão multidirecional do perímetro urbano. Os resultados mostraram que o aglomerado urbano como um todo, apresentou no período analisado, valores médios Dd (1,755) ou tendência de preenchimento das áreas centrais e diluição das periféricas; DPA (1,400) ou sinuosidade do perímetro urbano a medida em que estas áreas foram ocupadas e DPR (1,859) ou limitada organização espacial das estruturas urbanas. A análise fractal multidirecional de crescimento foi realizada em oito setores radiais de expansão urbana, cujos valores médios do parâmetro Dd no período 1905-2001 são abaixo mencionados: (I) Cantareira=1,829; (II) Zona Leste=1,898; (III) Mauá=1,848; (IV) Ipiranga=1,852; (V) Billings=1,858; (VI) Santo Amaro/Interlagos=1,848; (VII) Castelo Branco/Anhanguera=1,859 e (VIII) Anhanguera/Bandeirantes=1,863. Dentre os fatores que influenciaram o comportamento fractal destes setores, destaca-se a ocupação fragmentada pela topografia, hidrografia e restrições jurídico-ambientais nos setores I, III, IV, V, VI e VIII e, nos setores II, III, IV, VI e VII a maior densificação foi impulsionada por incentivos públicos e de infra-estrutura, que contribuíram para a concentração industrial e de serviços. / The aim of this research is to evaluate and apply fractal dimension analysis methodologies to study the growing of metropolitan areas, using thematic maps and satellite images of time series. The Sao Paulo metropolitan area it was choose as test area to calculate fractal dimension variability over 96 years time series (1905-2001). Fractal dimension of the metropolitan area as a whole and urban expansion radial sectors it was calculated using the following fractal dimension methods: occupation urban density (Dd); area-perimeter (DPA) and perimeter -scale- (DPR). The study was accomplished to thematic maps and orbital images from 1905 to 2001 that supplied information to estimate fractal dimensions from relations of density (Dd), areaperimeter and perimeter-scale (DPR). Fractal dimension of maps and images were to evaluate the time oscillation of empty urban spaces and the radial growing of urban perimeter. Over period analyzed, urban agglomeration as a whole show means values of Dd = 1.755 - a trend to fill central areas and dilution of occupation density at peripheral ones; DPA = 1.400 - related to tortuous urban boundary at time the urban peripheries were occupied; and DPR = 1.859 related to low spatial organization of urban structures. The radial growing fractal analysis was accomplished to eight urban radial sectors, whose show means values of Dd parameter over 1905-2001, as following:(I) Cantareira =1.829; (II) East Zone =1.898; (III) Máua=1.848; (IV) Ipiranga=1.852; (V) Billings=1.858; (VI) Santo Amaro/Interlagos =1.848; (VII) Castelo Branco /Anhanguera=1.859; e (VIII) Anhanguera/Bandeirantes =1.863. Among determinants factors on fractal behavior of sectors I, III, IV, V, VI e VIII, the results show the topographical, hydrographical and environment-legal constraints influenced the fragmented occupation and; for sectors II, III, IV, VI e VII... (Complete abstract, click eletronic address below). / Doutor

Submitted by Geyciane Santos (geyciane_thamires@hotmail.com) on 2015-12-15T20:29:07Z No. of bitstreams: 1 Dissertação - Vinicius de Moura Oliveira.pdf: 4829266 bytes, checksum: 8ff2ec4e1f047caf1c9b18b0c0923db3 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2016-01-20T18:40:26Z (GMT) No. of bitstreams: 1 Dissertação - Vinicius de Moura Oliveira.pdf: 4829266 bytes, checksum: 8ff2ec4e1f047caf1c9b18b0c0923db3 (MD5) / Approved for entry into archive by Divisão de Documentação/BC Biblioteca Central (ddbc@ufam.edu.br) on 2016-01-20T18:41:44Z (GMT) No. of bitstreams: 1 Dissertação - Vinicius de Moura Oliveira.pdf: 4829266 bytes, checksum: 8ff2ec4e1f047caf1c9b18b0c0923db3 (MD5) / Made available in DSpace on 2016-01-20T18:41:44Z (GMT). No. of bitstreams: 1 Dissertação - Vinicius de Moura Oliveira.pdf: 4829266 bytes, checksum: 8ff2ec4e1f047caf1c9b18b0c0923db3 (MD5) Previous issue date: 2015-09-29 / CAPES - Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The national energetic question has been a frequent schedule on headlines as for the final consumer cost, or for the cost of the environment. As it is approached to Santo Antonio´s hydropower, situated in the Rio Madeira, 7 km from Porto Velho, as the main factor of this dissertation it is intended to discuss the main brazilian energy matrix, its impacts in the landscapes and society around it. To instrumentalize this search, it has used the geoprocess as the initial data source and to analyse the effected environment by the hydropower enterprise. Having the data provided by the official departments, such as IBGE, INPE, ICMBIO and others, we traced a temporal space profile of usage and occupation on earth in the direct influence area at hydroelectric power plant (UHE) Santo Antonio. Then, we have elaborated thematic maps, tables and graphics which provided an indicative of evolution in the deforestation over the directed influenced area of the power plant, as the use given on earth in this period highlighted in this search specifically: from 2000 to 2012. As results of this search, it has observed a tendency of deforestation in the opposite senseof what has been expected, that is, it was noticed a decrease in the tax of deforestation in the area studied. This decrease represents about 20% of the deforestation in the beginning of the time which has reached 295 km² in 2000 to 60 km² in 2012. Most part of this area was destined for creating pasture, especially to the cattle. UHE Santo Antonio blunts as one of the main factors to stop the advance of deforestation in the area with its programs to recovery of degraded area and awareness of population around UHE. / A questão energética nacional tem sido pauta frequente do noticiário seja pelo seu custo ao consumidor final, seja pelo seu custo ao meio ambiente. Ao abordar a hidroelétrica Santo Antonio, situada no Rio Madeira, a 7 km de Porto Velho-RO,como agente principal desta dissertação buscou-se trazer à discussão a principal matriz energética brasileira, seus impactos na paisagem e na sociedade ao redor. Para instrumentalizar esta pesquisa, fez-se uso do geoprocessamento como principal fonte de dados e ferramenta de análise do ambiente afetado pelo empreendimento hidroelétrico. De posse dos dados disponibilizados por órgãos oficiais, tais como IBGE, INPE, ICMBio e outros, traçou-se um perfil espaço-temporal do uso e ocupação da terra na área de influência direta da Usina Hidroelétrica(UHE)Santo Antonio.Então, elaborou-se mapas temáticos, tabelas e gráficos que deram um indicativo da evolução do desflorestamento na área de influência direta da usina, bem como do uso dado à terra no período destacado da pesquisa, a saber: de 2000 a 2012. Como resultados da pesquisa verificou-se uma tendência de desflorestamento no sentido oposto ao esperado, isto é, percebeu-se um decréscimo na taxa de desmatamento na área estudada.Esse decréscimo representa cerca de 20% do desmatado no início da série que passou de 295 km² em 2000 para 60 km² em 2012. Grande parte desta área foi destinada a criação de pasto, especialmente para gado bovino. A UHE Santo Antonio desponta como um dos principais agentes a frear o avanço do desmatamento na área por seus programas de recuperação de área degradada e conscientização da população do entorno da UHE.

When scattering intensity is plotted against the dimensionless quantity qR, where q is the magnitude of the scattering wave vector and R is the radius of the particle, in log-log scale the scattering curve shows a power-law structure which defines characteristic crossovers. This work reveals some new relationships between the power-law structure and the particle properties. In this work, computer simulation results from T-matrix, Mie theory, and discrete dipole approximation methods are used to study the far field intensity and the internal field of the particles. Scattering by both weakly and strongly refractive particles are studied. For weakly refractive randomly oriented spheroidal particles, how the phasor cancellation-based tip volume method can be applied to predict the scattering envelope is demonstrated. The relationship between backscattering enhancement and the curvature of the weakly refractive particles is explained. In strongly-refractive particles when the phase shift parameter is high, regions with higher field amplitudes start to appear. These regions are recognized as the hot spot regions. In this work, a proper definition is given to the hot spot region. The relationships between the hot spot region and the power-law structure, between the hot spot region and the particle morphology, and between the power-law structure and the particle morphology are extensively studied for scattering by spherical particles. A new semi-quantitative phasor analysis method is introduced, and the new method is used with color-coded phasor plots to identify how different regions of the particle contribute to the scattering pattern to get an insight into the physics behind the scattering. How different regions of the particle contribute to the second crossover (SC) and the backscattering enhancement is presented. Relationships between the SC, particle size, and relative refractive index of the particle are derived. It was identified that the scattering angle at the SC depends only on the relative refractive index of the particle. How the findings of this work can be applied to solve the inverse electromagnetic scattering problem for a single non-absorbing spherical particle is also discussed.

<p>In this dissertation, numerical analysis frameworks and software tools for digital design of process systems are developed. More specifically, these tools have been focused on digital design within the pharmaceutical manufacturing space. Batch processing represents the traditional and still predominant pathway to manufacture pharmaceuticals in both the drug substance and drug product spaces. Drug substance processes start with raw materials or precursors to produce an active pharmaceutical ingredient (API) through synthesis and purification. Drug product processes take this pure API in powder form, add excipients, and process the powder into consumer doses such as capsules or tablets.  Continuous manufacturing has allowed many other chemical industries to take advantage of real-time process management through process control, process optimization, and real-time detection of off-spec material. Also, the possibility to reduce total cleaning time of units and encourage green chemistry through solvent reduction or recycling make continuous manufacturing an attractive alternative to batch manufacturing. However, to fully understand and take advantage of real-time process management, digital tools are required, both as soft sensors during process control or during process design and optimization.  Since the shift from batch to continuous manufacturing will proceed in stages, processes will likely adopt both continuous and batch unit operations in the same process, which we will call {\em hybrid} pharmaceutical manufacturing routes. Even though these processes will soon become common in the industry, digital tools that address comparison of batch, hybrid, and continuous manufacturing routes in the pharmaceutical space are lacking. This is especially true when considering hybrid routes. For this reason, PharmaPy, an open-source tool for pharmaceutical process development, was created to address rapid in-silico design of hybrid pharmaceutical processes.  Throughout this work, the focus is on analyzing alternative operating modes within the drug substance manufacturing context. First, the mathematical models for PharmaPy's synthesis, crystallization, and filtration units are discussed. Then, the simulation capabilities of PharmaPy are highlighted, showcasing dynamic simulation of both fully continuous and hybrid processes. However, the technical focus of the work as a whole is primarily on optimization techniques for pharmaceutical process design. Thus, many derivative-free optimization frameworks for simulation-optimization were constructed and utilized with PharmaPy performing simulations of pharmaceutical processes.  The timeline of work originally began with derivative-based methods to solve mixed-integer programs (MIP) for water network sampling and security, as well as nonlinear programs (NLPs) and some mixed-integer nonlinear programs (MINLPs) for design space and feasibility analysis. Therefore, a method for process design that combines both the ease of implementation from a process simulator (PharmaPy) with the computational performance of derivative-based optimization was implemented. Recent developments in Pyomo through the PyNumero package allow callbacks to an input-output or black-box model while using {\sc Ipopt} as a derivative-based solver through the cyipopt interface. Using this approach, it was found that using a PharmaPy simulation as a black box within a derivative-based solver resulted in quicker solve times when compared with traditional derivative-free optimization strategies, and offers a much quicker implementation strategy than using a simultaneous equation-oriented algebraic definition of the problem.  Also, uncertainty exists in virtually all process systems. Traditionally, uncertainty is analyzed through sampling approaches such as Monte Carlo simulation. These sampling approaches quickly become computational obstacles as problem scale increases. In the 1980s, chemical plant design under uncertainty through {\em flexibility analysis} became an option for explicitly considering model uncertainty using mathematical programming. However, such formulations provide computational obstacles of their own as most process models produce challenging MINLPs under the flexibility analysis framework.  Specifically when considering pharmaceutical processes, recent initiatives by the FDA have peaked interest in flexibility analysis because of the so called {\em design space}. The design space is the region for which critical quality attributes (CQAs) may be guaranteed over a set of interactions between the inputs and process parameters. Since uncertainty is intrinsic to such operations, industry is interested in guaranteeing that CQAs hold with a set confidence level over a given operating region. In this work, the {\em probabilistic design space} defined by these levels of confidence is presented to address the computational advantages of using a fully model-based flexibility analysis framework instead of a Monte Carlo sampling approach. From the results, it is seen that using the flexibility analysis framework decreased design space identification time by more than two orders of magnitude.  Given implementation difficulty with new digital tools for both students and professionals, educational material was developed for PharmaPy and was presented as part of a pharmaceutical API process development course at Purdue. The students were surveyed afterward and many of the students found the framework to be approachable through the use of Jupyter notebooks, and would consider using PharmaPy and Python for pharmaceutical modeling and data analysis in the future, respectively.  Through software development and the development of numerical analysis frameworks, digital design of pharmaceutical processes has expanded and become more approachable. The incorporation of rigorous simulations under process uncertainty promotes the use of digital tools in regulatory filings and reduces unnecessary process development costs using model-based design. Examples of these improvements are evident through the development of PharmaPy, a simulation-optimization framework using PharmaPy, and flexibility analysis tools. These tools resulted in a computational benefit of 1 to 2 orders of magnitude when compared to methods used in practice and in some cases reduce the modeling time required to determine optimal operating conditions, or the design space of a pharmaceutical manufacturing process.</p>