A tese propõem uma metodologia, em nível de pesquisa, por intermédio do desenvolvimento e da adaptação de ferramentas de apoio computadorizado, capaz de realizar a segmentação da imagem dos discos intervertebrais da região lombar da coluna vertebral humana, de maneira semiautomática reduzindo drasticamente o tempo gasto manualmente neste procedimento, sem perder sua acurácia e, ainda, garantindo maior reprodutibilidade em seus resultados. Foram utilizadas imagens sagitais de ressonância magnética ponderadas em T2 de 285 discos intervertebrais de 70 pacientes, classificados segundo o grau de severidade da degeneração discal definido pelo critério proposto por Pfirrmann. A classificação computacional dos discos foi realizada com base em atributos quantitativos extraídos dos histogramas de níveis de cinza e de informações de textura das imagens. O desempenho dos métodos computacionais de segmentação foi avaliado com base no Coeficiente de Jaccard, na distância de Hausdorff e no Erro Médio Quadrático. O desempenho dos métodos computacionais de classificação foi também avaliado com base em medidas similares à aplicação da sensibilidade, da especificidade e da área sob a curva ROC. A segmentação manual e a classificação por inspeção visual dos discos realizadas por três profissionais experientes foram utilizadas como padrão ouro para a comparação. Os principais resultados indicaram a médio de 63,22% para o Coeficiente de Jaccard, as médias de 0,044 das distâncias de Hausdoff e de 0,014 para o EMQ na comparação entre as imagens. Além disso, a segmentação semiautomatizada diferiu em uma taxa média de 30% em relação à segmentação manual e a classificação da degeneração discal, por redes neurais artificiais difere em menos de 2%, ao ser comparada ao procedimento de classificação manual realizado pelos especialistas. / The thesis proposes a methodology at the level of research through the development and adaptation of computerized support tools, able to perform the image segmentation of the intervertebral discs of the lumbar region of the human spine, semiautomatic way dramatically reducing time spent manually in this procedure, without losing its accuracy and also ensuring more reproducible in their results. Were used sagittal MRI T2- weighted of 285 intervertebral discs from 70 patients, classified according to the severity of disc degeneration defined by the criteria proposed by Pfirrmann. The computational classification of disks was based on quantitative attributes extracted from histograms of gray level images and the texture information. The performance of computational segmentation methods was evaluated based on Jaccard coefficient, Hausdorff distance and Mean Square Error. The performance of the computational classification methods was evaluated based on measures of sensitivity, specificity and the area under the ROC curve. The manual segmentation and visual inspection classification of the discs made by three experienced professionals were used as the gold standard for comparison. The main results showed an average Jaccard coefficient of 63.22%, the average Hausdoff of distances was 0.044 and 0.014 Mean Square Error average when comparing the images from both segmentation targets. Additionally, the targeting semiautomatic differed by an average of 30% compared with manual segmentation and classification of disc degeneration provided from an artificial neural networks differs by less than 2% when compared to manual sorting procedure performed by experts.

Submitted by Silvana Teresinha Dornelles Studzinski (sstudzinski) on 2016-11-03T11:54:04Z No. of bitstreams: 1 Murillo da Silveira Grübler_.pdf: 3027339 bytes, checksum: 2fdb175c76ab90e275bf8ba04792e452 (MD5) / Made available in DSpace on 2016-11-03T11:54:04Z (GMT). No. of bitstreams: 1 Murillo da Silveira Grübler_.pdf: 3027339 bytes, checksum: 2fdb175c76ab90e275bf8ba04792e452 (MD5) Previous issue date: 2016-08-19 / CNPQ – Conselho Nacional de Desenvolvimento Científico e Tecnológico / O Gerenciamento de Leitos é uma importante área de planejamento e controle hospitalar. Sua função é garantir o equilíbrio entre os pacientes que chegam através do setor de emergência, os eletivos que possuem algum tratamento agendado e aqueles que saem do hospital. Dessa forma, esse gerenciamento possibilita manter alta a taxa de ocupação dos quartos, mas sem realmente lotá-los, além de prever qualquer situação não planejada. A gestão eficaz de leitos hospitalares como recurso sempre foi um desafio para os gestores. Nas décadas de 80 e 90, por exemplo, milhares de pacientes tiveram operações canceladas em virtude de razões não médicas. Como há necessidade de um melhor controle do fluxo, a área de Gerenciamento de Leitos começou, então, a receber mais atenção acadêmica e também políticas nacionais para a sua gestão. O processo de admissão e posicionamento de enfermos, a partir do Gerenciamento de Leitos, vem se desenvolvendo nos últimos anos através de diversas técnicas de pesquisa operacional, tais como simulação, teoria de filas, análise estatística, entre outras. Devido às constantes incertezas vividas pelos hospitais atualmente, o uso do modelo cognitivo Ciência de Situação em pesquisas científicas na área da saúde vem crescendo cada vez mais. A Ciência de Situação é uma área de estudo que busca compreender o contexto dos ambientes e projetar ações futura. Em suma, é uma técnica que vai além do tradicional processamento de informações, visto que procura explicar o comportamento humano na operação de sistemas complexos. Nessa assertiva, este trabalho tem como objetivo utilizar a Ciência de Situação na área de Gerenciamento de Leitos, usando um modelo híbrido que une a técnica de Rede Neural Artificial Multilayer Perceptron com a Teoria do Valor Multiatributo para tomada de decisão, auxiliando gestores no processo de atribuição de pacientes em leitos adequados ao seu tratamento. Através da implementação de um protótipo baseado neste modelo híbrido de apoio à decisão, nomeado de IMBEDS, foram avaliados 50 pacientes em um total de 266 leitos gerenciados pela Central de Leitos, no Hospital Mãe de Deus, localizado em Porto Alegre. O resultado final dos testes foi de 93,5% de similaridade entre o leito apto apresentado pelo modelo e o processo real de alocação dos enfermos. / The Bed Management is an important area of planning and control hospital. It’s function is to ensure the balance between the patients who come through the emergency department, elective that have some scheduled treatment and those leaving the hospital. Thus, the Bed Management enables the hospital keep high occupancy rate of rooms, but without fill all the beds, in addition to providing any unplanned situation. Effective management of hospital beds as a resource has always been a challenge for managers. In the 80s and 90s, for example, thousands of patients have operations canceled due to non-medical reasons. As there is need for better control of the flow, Bed Management area then began to receive more academic attention and also policies national for the Bed Management. The process of admission and positioning the patients, from the management of beds, has been developing in recent years through of operational research, such as simulation, queuing theory, statistical analysis, among others. Due to the uncertainties experienced by hospitals nowadays, the use of model Situation Awareness in research in the health field is growing increasingly. Situation Awareness is a field of study that seeks to understand the context of the environment and designing future actions. In short, it is a technique that goes beyond the traditional information processing, as it seeks to explain human behavior in the operation of complex systems. In this statement, this work aims to use the Situation Awareness in Bed Management area, using a hybrid model that combines the technique Artificial Neural Network Multilayer Perceptron with the Multi-Attribute Value Theory for decision making, assisting managers in process of patient's allocation to the bed suitable in his treatment. Through the implementation of a prototype based on this hybrid model of decision support, named of IMBEDS, were evaluated 50 patients in a total of 266 beds managed by Beds Center, in the Hospital Mãe de Deus, located in Porto Alegre. The final result of the tests was 93.5% similarity between the bed apt selected by the model and the allocation process of the patients.

Numa área experimental de 25,2 ha formada com o capim-braquiarão (Brachiaria brizantha (Hochst) Stapf.) cv. Marandu e localizada no Campus da USP em Pirassununga/SP, durante o período de janeiro a julho de 2004, conduziu-se a presente pesquisa pela Faculdade de Zootecnia e Engenharia de Alimentos (FZEA/USP) com os seguintes objetivos: 1) Determinar as frações de carboidratos (A - açúcares solúveis com rápida degradação ruminal; B1- amido e pectina; B2 - parede celular com taxa de degradação mais lenta; C - fração não digerida) e as frações protéicas (A - NNP; B1 - peptídeos e oligopeptídeos; B2 - proteína verdadeira; B3 - NFDN; C - NIDA) na forragem da gramínea, baseados nas equações utilizadas pelo método de Cornell; 2) Relacionar outras variáveis com as medições em campo de experimentos paralelos e dados de elementos de clima com as frações protéicas e de carboidratos com o auxílio de um modelo computacional baseado em redes neurais artificiais (RNA). O delineamento foi em blocos completos e casualizados, com quatro tratamentos (ofertas de forragem de 5, 10, 15 e 20% - kg de massa seca por 100 kg de peso animal.dia) e quatro repetições. Cada bloco era dividido em quatro unidades experimentais de 1,575 ha, com cinco piquetes de 0,315 ha cada. Os animais eram manejados em cada unidade experimental em lotação rotacionada, com períodos de descanso de 28 dias no verão e 56 dias no inverno e período de ocupação de 7 dias, respectivamente. As amostras eram colhidas 2 dias antes da entrada dos animais à altura do resíduo do pastejo anterior. Foram determinados produção de massa seca (MS), alturas de pré e pós pastejo, fibras em detergente ácido (FDA) e neutro (FDN), sacarose, amido, lignina, extrato etéro (EE), carboidrato totais (CHO), carboidratos não estruturais (CNE), frações A, B1, B2 e C de carboidratos, proteína bruta (PB), frações A, B1, B2, B3 e C de proteínas e análise de uma rede neural artificial para uma predição dos teores de FDA, fibra em detergente neutro, PB e as frações protéicas. A produção de massa seca (MS) foi significativa, quando se estudou os efeitos da oferta de forragem (p<0,05), ciclo de pastejo (p<0,05) e da interação oferta de forragem x ciclo de pastejo (p<0,05). A maior produção foi no mês de março, quando se alcançou a média de 16140 kg MS/há para o oferta de 20%. Os teores de FDA foram significativos, quando se estudou a oferta de forragem (p<0,05) e seus maiores. Os teores médios da fibra em detergente neutro foram de 66,3 e 64,7% no verão e inverno respectivamente. Houve diferenças significativas para PB, quando se estudou a oferta de forragem (p<0,05), sendo seus teores médios de maior valor na OF a 5%. Observa aumento dos CNE em função de lâminas e colmos ao longo das estações do ano com interação no CP x OF (p<0,05) e seus maiores valores foram encontrados no ciclo de pastejo 3 na oferta de forragem 5%. Os teores de CHO totais apresentaram diferenças (p<0,10) em função da oferta de forragem, sendo os maiores teores médios encontrados na oferta de forragem de 20%. As frações A e B2 de CHO foram significativas em função da oferta de forragem (p<0,05), enquanto que os maiores teores médios da fração A foram encontrados nos ciclos de pastejo 3 e 4 e das frações B2 (%CHO) no ciclo de pastejo 1. As frações B2 e C de CHO apresentaram-se diferentes (p<0,05) nos ciclos de pastejo, sendo decrescentes para a fração B1 e crescentes para a fração C. As frações A (47%), B1 (11%) e B3 (10%) de proteínas foram significativas nos ciclos de pastejos. Os teores médios da fração B2 de proteínas apresentaram-se semelhantes (p>0,05) e os da fração C de proteínas foram diferentes (p<0,05) nas ofertas de forragem e ciclos de pastejo. Conclui-se que os ciclos de pastejos interferiram em todas as variáveis estudadas e que os teores das frações de proteínas e carboidratos estão dentro da variação (%) encontrada na literatura. A rede neural artificial conseguiu vincular as interações existentes de dados de campo e estimar os valores laboratoriais dentro de erros esperados, permitindo com isso desvincular análises laboratoriais, de qualidade de planta forrageira, à pesquisa agropecuária e com isso obter além de resultados mais rápidos, menor custo de pesquisa. / In a experimental área of 25.2 há formed with capim-braquiarão (Brachiaria brizantha (Hochst) Stapf ) cv. Marandu located in University of São Paulo Campus of Pirassununga/SP, during the period of january to july of 2004 was lead the present recherché for Faculdade de Zootecnia e Engenharia de Alimentos (FZEA/USP) to appetent the following objectives: 1) Determine protein fractions (the NNP; B1 - peptides and oligopepitides; B2 - true protein; B3 - NDF, C - AND) and carbohydrates fractions (soluble sugars with fast rumem degradation); B1(starch and pectin); B2 (cell wall alower degradation rate; C (indigested fraction rate) in the fodder plant of the grass, as it\'s respetive dregadability rate, based on equations using Cornell model. 2) To relate other variables measurements in field to parallel experiments and climate elements to the protein and carbohydrate fractions was used a computacional model based in nets of artificial neural. The randomized complete block design with four treatments (herbage allowance of 5, 10, 15 and 20% - kg of dry mass for 100 kg of animal.dia weight) and four repetitions. Each block was divided in four experimental units of 1,575 ha, with five 0,315 poles of ha each. The animals were management in each experimental unit in rotational grazing capacity, with periods of rest of 28 days in the summer and 56 days in the winter and period of occupation of 7 days, respectively. The samples were harvested 2 days before the entrance of the animals to the height of the residue of pasture previous. Were conducted analysis of production of dry mass (DM), heights daily pay and after grazing, staple fibers in acid detergent (ADF) and neutral (NDF), sacarose, starch, lignina, extract etereo (EE), carbohydrate (CHO), not structural carbohydrate (NSC), fractions A, B1, B2 and C of carbohydrate, crude protein (CP), fractions protein A, B1, B2, B3 and C and analysis of artificial neural network for a prediction of levels of ADF, NDF, CP and protéicas fractions. The dry matter (DM) production was significant for herbage allowance (p<0,05), grazing periods (p<0,05) and interaction between allowances x grazing periods (p<0,05). The righ production was in February 13,352 kg MS/ha. The ADF was significant for allowance and grazing periods (p<0,05), with 34.8%, on summer and 35.9% on winter. The average measured of NDF on summer and winter was 66.3 and 64.7%, respectively. It showed significant differences of PC when studied the allowance (p<0,05) and its average measured on summer and winter was 8,3 and 8,1%, respectively. It observes increase of the CNE in function of blades and stem to the long one of the stations of the year with interaction in grazing periods x herbage allowance and its bigger values had been found in the grazing periods 3 with herbage allowance 5%. The total texts of CHO had presented differences (p<0,10) in function of herbage allowance, being biggest found average texts in herbage allowance of 20%. The fractions and the B2 of CHO had been significant, when studied in function of the herbage allowance (p<0,05) for the fraction A and for fraction B2 (p<0,05); the biggest average texts in % of CHO of the fraction had been found It in the cycles of grazing 3 and 4 and the B2 fractions (%CHO) in the grazing periods 1. Fractions B2 (p<0,05) and C (p<0,05) of CHO had presented significant differences, when studied the factor grazing periods, where the B1 fraction the texts had been diminishing the measure that increased the grazing periods and fraction C the texts had increased the measure that had increased the grazing periods. The A, B1 and B3 protein fraction was significant when was studied the grazing periods and the results were 0,47; 0,11; 0,10 respectively. The B2 fraction was not significant. C fraction was significant when studied the allowance (p<0,05) and grazing periods (p<0,05). It was concluded that the grazing periods had intervened with all the studied 0 variable and that the texts of the protein fractions and carbohydrates are inside of the variation (%) found in literature. The results from lab was used to train and test neural network. With a program developed by neural network in a mult layer perceptron with capacity to predict the parameters of nutrition and nourishing value from parameters of forage plant intrinsic and extrinsic, where it was allowed to disentail lab analysis of forage plant quality on the farm research, to get beyond faster and have less research costs.

L'intérêt pour les véhicules électriques ne cesse de croitre au sein de la société contemporaine compte tenu de ses nombreuses interrogations sur l’environnement et la dépendance énergétique. Dans ce travail de thèse, nous essayons d’améliorer l’acceptabtabilité sociétale du véhicule électrique en essayant de faire avancer la recherche sur le diagnostique des défauts d’une chaine de traction électrique. Les résultats escomptés devraient permettre à terme d’améliorer la fiabilité et la durabilité de ces systèmes.Nous commençons par une revue des problèmes des défauts déjà apparus dans les véhicules hybrides séries qui disposent de l’architecture la plus proche du véhicule électrique. Une étude approfondie sur le diagnostic des défauts d’un convertisseur de puissance statique (AC-DC) ainsi que celle du moteur synchrone à aimants permanents est menée. Quatre types de défauts majeurs ont été répertoriés concernant le moteur (court-circuit au stator, démagnétisation, excentricité du rotor et défaut des roulements). Au niveau du convertisseur, nous avons considéré le défaut d’ouverture des interrupteurs. Afin d’être dans les mêmes conditions d’utilisation réelle, nous avons effectué des tests expérimentaux à vitesse et charge variables. Ce travail est basé aussi bien sur l’expérimentation que sur la modélisation. Comme par exemple, la méthode des éléments finis pour l’étude de la démagnétisation de la machine. De même, l’essai en court-circuit du stator du moteur en présence d’un contrôle vectoriel.Afin de réaliser un diagnostic en ligne des défauts, nous avons développé un modèle basé sur les réseaux de neurones. L’apprentissage de ce réseau de neurone a été effectué sur la base des résultats expérimentaux et de simulations, que nous avons réalisées. Le réseau de neurones est capable d'assimiler beaucoup de données. Ceci nous permet de classifier les défauts en termes de sévérité et de les localiser. Il permet ainsi d'évaluer le degré de performance de la chaine de traction électrique en ligne en présence des défauts et nous renseigner ainsi sur l'état de santé du système. Ces résultats devraient aboutir à l’élaboration d’une stratégie de contrôle tolérant aux défauts auto-reconfigurable pour prendre en compte les modes dégradés permettant une continuité de service du véhicule ce qui améliorera sa disponibilité. / The interest in the electric vehicles rose recently due both to environmental questions and to energetic dependence of the contemporary society. Accordingly, it is necessary to study and implement in these vehicle fault diagnosis systems which enable them to be more reliable and safe enhancing its sustainability. In this work after a review on problem of faults in the drivetrain of series hybrid electric vehicles (SHEV), a deep investigation on fault diagnosis of AC-DC power converter and permanent magnet synchronous motor (PMSM) have been done as two important parts of traction chains in SHEVs. In other major part of this work, four types of faults (stator winding inter turn short circuit, demagnetization, eccentricity ant bearing faults) of a PMSM have been studied. Inter turn short circuit of stator winding of PMSM in different speeds and loads has been considered to identify fault feature in all operation aspects, as it is expected by electric vehicle application. Experimental results aiming short circuits, bearing and eccentricity fault detection has been presented. Analytical and finite element method (FEM) aiming demagnetization fault investigation has been developed. The AC-DC converter switches are generally exposed to the possibility of outbreak open phase faults because of troubles of the switching devices. This work proposes a robust and efficient identification method for data acquisition selection aiming fault analysis and detection. Two new patterns under AC-DC converter failure are identified and presented. To achieve this goal, four different level of switches fault are considered on the basis of both simulation and experimental results. For accuracy needs of the identified pattern for SHEV application, several parameters have been considered namely: capacitor size changes, load and speed variations. On the basis of the developed fault sensitive models above, an ANN based fault detection, diagnosis strategy and the related algorithm have been developed to show the way of using the identified patterns in the supervision and the diagnosis of the PMSM drivetrain of SHEVs. ANN method have been used to develop three diagnosis based models for : the vector controlled PMSM under inter turn short circuit, the AC/DC power converter under an open phase fault and also the PMSM under unbalanced voltage caused by open phase DC/AC inverter. These models allow supervising the main components of the PMSM drivetrains used to propel the SHEV. The ANN advantages of ability to include a lot of data mad possible to classify the faults in terms of their type and severity. This allows estimating the performance degree of that drivetrains during faulty conditions through the parameter state of health (SOH). The latter can be used in a global control strategy of PMSM control in degraded mode in which the control is auto-adjusted when a defect occurs on the system. The goal is to ensure a continuity of service of the SHEV in faulty conditions to improve its reliability.

Orientador: Carlos Roberto Minussi / Resumo: Nesta pesquisa de doutorado, propõe-se o desenvolvimento de um sistema previsor de carga multinodal, via Redes Neurais Artificiais (RNAs). Trata-se da previsão que envolve vários pontos da rede elétrica, e.g., subestações de sistemas de distribuição, alimentadores, transformadores etc., bem como as barras dos sistemas de transmissão de grande porte. Na literatura especializada, há a prevalência de oferta de propostas visando à previsão da carga total (ou global) correspondente ao somatório de todo o consumo demandado no sistema, considerando-se um horizonte, por exemplo, 24 horas à frente. Nesta pesquisa, dar-se-á ênfase à previsão de carga multinodal. Visando realizar esta previsão, há necessidade de se dispor de um procedimento especializado que produza resultados que atendam os requisitos do setor elétrico (precisão desejada, confiabilidade e rapidez). Estes requisitos são os objetivos desta pesquisa, cujo modelo desenvolvido constitui-se num sistema neural inspirado na arquitetura neural da família ART (Adaptive Resonance Theory), mais especificamente, a RNA supervisionada ARTMAP-Fuzzy, a qual congrega a teoria da ressonância adaptativa e a teoria dos conjuntos fuzzy. O emprego da teoria dos conjuntos fuzzy confere, às RNAs da família ART, a aptidão de processar informações analógicas, binárias, assim como combinações dessas informações. A opção por esta RNA é em razão do seu atributo de ser estável e plástica. A estabilidade está associada à capacidade de produzir sempre... (Resumo completo, clicar acesso eletrônico abaixo) / Abstract: This research proposes to develop a multinodal load forecasting system by Artificial Neural Networks. This is a prediction for several points of the electrical network, e.g. distribution substations, feeders, transformers, etc., as well as busses of huge transmission systems. The literature offers proposals for total load forecasting (or global) corresponding to the sum of all demanded consumption considering a horizon of 24h ahead. This research emphasizes multinodal load forecasting. To perform this prediction, it is necessary to have a specialized procedure that provides results attending the requests of the electrical system (desired precision, reliability and velocity). These requests are the objective of this research, whose developed model is based on ART (Adaptive Resonance Theory) family, specifically the supervised Fuzzy ARTMAP neural network that uses the adaptive resonance theory and fuzzy logic theory. The option of this neural network is due to the attribute to be stable and plastic. The stability is associated to the capacity to produce always a solution. The plasticity (incremental training) is a propriety that is not observed in most of the neural network available on the literature. This is similar to what occurs with humans, as new information comes, the human being is more intelligent. Knowing the electrical load with precision and in advance is a primordial need. The studies about the operational modes of the system and the strategies used to attend conti... (Complete abstract click electronic access below) / Doutor

在資訊爆炸的時代，處於日趨複雜的環境及多重資訊來源管道之下，如何從大量及瑣碎的資訊中找出「重要且有用」的部份，藉以輔助企業或個人制定正確的決策，並降低資訊取得的成本，是資訊人員在設計資訊系統時所必須考量的重要因素之一，因此，資訊篩選(Information filtering)已成為當務之急，更顯示出其重要性。 本研究之主要目的在於整合類神經網路與模糊理論以建立一個通用型資訊篩選之演算法，藉由此演算法可篩選出重要之決策變數，減少資訊的使用量，達到相同或類似的決策結果，進而降低後續資訊蒐集之成本。最後並以四個XOR實驗及國內上市公訂股價預測為例，以測試本研究所開發出來之演算法的正確性及實用性。就XOR實驗結果顯示均能迅速且正確的篩選出重要的輸入資訊；而在股價預測方面，結合基本面分析及技術面分析，根據個別公司的特性及不同的時間點，能夠篩選出其重要的預測變數，可作為股市投資者之重要參考依據。因此，藉由本演算法所開發出來的系統，可以達到資訊篩選的目的。 / At the time of information explosion, how to filter the important and useful parts from a large and trivial information pool is one of the most important factors considering in designing information systems which are used to assist users making right decisions by MIS managers. The purpose of this research is to integrate two technologies. Artificial Neural Network and Fuzzy Theory, to develop a generalized algorithm to filter important information. We hope that using this algorithm we can (1)filter the important decision variables, (2)decrease the information usage, and (3)reduce the cost of information collection. Finally, we made four experiments on the XOR system and stock market forecasting to test the accuracy and practicability of the information filter algorithm. The results of experiments showed that the algorithm could filter the important information correctly and quickly.

財務報表記錄可說是企業經營績效良窳的反映指標，而其中所衍生出來的財務比率，向 來均是管理者、投資者進行企業診斷或未來經營績效預測的重要資訊來源。然而，相關 的研究發現，由於產業間經濟環境與市場結構特性的不同，所呈現出來的財務報表資訊 內涵亦將有所差別。因此，若進一步運用個別產業之報表資訊預測公司未來盈餘時，將 能夠提供產業間結果進行分析與比較的基礎。 如何自報表中獲取與公司經營績效相關之會計資訊，進而建構出優良的盈餘預測模式， 是近幾年來學者感興趣的研究課題之一。鑑於人工智慧之類神經網路系統擁有多項的特點，因此，對於盈餘預測會計資訊萃取的應用上，無非是提供了我們一個新的選擇途徑。 本研究即根據此項概念，以民國70年第一季至民國82年第三季為止共十五項大小產業之 股票上市公司財務報表以及股價報酬等資料作為研究樣本，進行盈餘預測模式的建構以 及投資超額報酬的計算。 進一步地說，本研究的內容可以分成三個部份，第一部份是以整體市場樣本為例，對類 神經網路主要參數如輸入變數組合、隱藏層節點數等進行調整及測試，以從中選取出盈 餘預測效果較佳之模式設定；在第二部份則是運用此一盈餘預測模式，分別對整體市場 以及紡織、電子電機 兩項產業樣本進行網路的訓練與測試，並根據模式所獲得之區別及 預測能力評估指標，探討不同產業特性樣本所建構的模式之間，其預測結果上的差異性 ；而第三部份則是利用各類產業模式預測結果的資訊，從利潤與風險兩種角度，定義"總 體"、"高利潤"、"低風險"、 "高利潤低風險"等四種不同類型投資策略，並以事件研究 法計算各項策略所能獲取之累積超額報酬，最後，則根據各策略之獲利績效，進行產業 間的分析比較，以找出本研究各類特定產業之最適投資策略。 本研究根據前述方式所進行的實驗研究中，獲得了以下三點結論： 一、類神經網路盈餘預測模式之建構 (一)以整體市場樣本為對象所進行之網路的測試中，發現模式整體區別能力大致介於五 到七成之間；而整體預測能力則介於四到六成之間。 (二)本研究所找出盈餘預測效果較佳之網路模式設定如下：1.輸入變數組合：單因子多變量變異數分析之22項顯著性財務比率 2.網路架構(輸入層-隱藏層-輸出層)：22-22-1 3.連結權數初始值設定範圍：-0.1~0.1 二、產業盈餘預測結果之分析 (一)整體而言，產業間模式測試結果的差異並不大，其中以紡織產業的模式區別及預測 能力最好(70%以上)，電子電機產業次之，而整體市場模式的結果均不及兩項單一性產業。 (二)模式預測能力穩定性方面，各產業於五個年度間預測率的波動大致還算穩定，其中 就紡織產業而言，其年度之間模式預測能力的差別不大，但電子電機產業年度間的變化 則要比前者來得明顯。 三、產業投資策略績效之分析 (一)各類型投資策略的整體結果中，紡織與電子電機兩項產業的獲利績效相當，且均要 比整體市場來得好，其中，紡織產業之"高利潤低風險"策略所獲得的累積超額報酬(43.28%) 更居全體之冠。 (二)本研究所找出之個別產業最適投資策略分別為： 1.整體市場：總體策略、低風險策略 2.紡織產業：高利潤低風險策略、高利潤策略 3.電子電機產業：高利潤低風險策略、低風險策略 / Financial Statements are very important information indicating performance of corporations. Managers and investors use financial ratios as vital indexes to evaluate and predict operating results of corporations, and make their decisions. ategy, and compute CAR for each investment strategies. At last, I analyze the investing results of the four strategies for individual industry. ANN ( Artificial Nerual Network) shoot a new direction on researching application of abstracting accounting information which can efficiently predict earnings. According to results of relative researches, financial statements from different industries present and implicate different accounting information. If we further apply ANN on financial statement information to predict earnings of corporations, we can use the results as bases of analyses and comparisons among industries. Because ANN model has many advantages, in this research, I use financial statements and return on stocks from corporations as researching samples to construct prediction models and compute CAR(Cumulative Abcdrmal Return) on investments. These samples are chosen from 15 different industries and covered from the first quarter of 1981 to the third quarter of 1993. This research consists of three parts: 22 financial ratios selected by MANOVA First, I use the general market samples to adjust and predict the vital parameters of ANN models, such as the selection of input variable, the number of hidden node, and finally pick better setups for the prediction model. Second, I use this model to train and test samples from the general market, the textile, and the electron industry, and research the variation of predicting results by different models made up different industries by means of evaluation indexes . Third, I use the results predicted by the three different industry models, inspect of risk and return, to define four types of investment strategies -- "the general", "the high return", "the low risk", and "the high return - low risk" strategy, and compute CAR for each investment strategies. At last, I analyze the investing results of the four strategies for individual industry. After researching, I find:s of the textile and electron industry are better than the general markets'. 1.The better setups of ANN predition models are :industries are: (1)the selection of input variable:the 22 financial ratios selected by MANOVA (2)the ANN model topology(input node - hidden node - output node):22-22-1 rategy (3)the range of initial connection weights:-0.1~0.1 return - low risk strategy 2.The analyses of results predicted by the three different industry models are: (1)the predicting abilities of the textile and electron industry are better than the general markets'. 3.The proper investment strategies of individual industries are: (1)the general market:the general and the low risk strategy (2)the textile industry:the high return and the high return - low risk strategy (3)the electron industry:the low risk and the high return - low risk strategy

GPS 單點定位或稱絕對定位，傳統上使用虛擬距離觀測量，容易受到 電離層延遲影響，導致定位精度較差。因此，本文的目的為構建即時的區 域性電離層模型，以便能夠即時減弱電離層延遲量，提高單頻GPS 單點定 位的精度。 構建電離層模型的方法有很多種，而運用類神經網路為可能方法之一， 但是， 國內較少人探討。本研究嘗詴使用倒傳遞類神經網路(Back-propagation Artificial Neural Network)，構建即時的區域電離層模型，藉由選擇適當的神經訓練函數及隱藏層神經元，利用過去收集的已知參考站的雙頻GPS 資料，計算電離層延遲量，訓練類神經網路，直到精度合乎要求；再以檢核站GPS 資料，檢驗類神經網路預測電離層延遲的功效。 採用的實驗資料為臺南市政府e-GPS 系統所提供六個測站，2008 年1 月3 日到1 月5 日的GPS 資料，計算測站與GPS 衛星連線中假想的電離層 薄殼交點—電離層穿透點(Ionosphere Pierce Point, IPP）之地理位置(緯度φ、經度λ)，及太陽黑子數(sunspot numbers)等當作輸入值，IPP 的垂直電離層延遲當作輸出值，測詴包含單日、兩日以及不同的資料型態(IPP 點、網格點)等情況訓練類神經網路，藉由相對應的驗證資料，檢驗類神經網路的功效，最後將類神經網路的預估成果與全球電離層改正模型、雙頻GPS 資料計算的電離層延遲相比較，並根據改正率與統計特性，評估類神經網 路構建出的區域性電離層模型的成效。 由實驗成果顯示，構建的即時區域性電離層模型的標準差可小於±3TECU，並可改正約80%的電離層延遲誤差，故以類神經網路可有效的構 建出區域性的電離層模型。 / The conventional single point positioning using GPS pseudo rangemeasurements, are vulnerable to ionospheric errors, leading to poor positioningaccuracy. Constructing a real-time ionospheric model is one of the methods that can reduce the ionospheric errors and improve the single point positioning accuracy. Although there are many methods to construct regional ionosphere model,using artificial neural network (ANN) to construct a real-time ionospheric model is less to be mentioned. This study used back-propagation artificial neural network to estimate a regional real-time ionospheric model by selecting the appropriate training functions and the number of hidden layers and its’ nodes. The neural network had to be ‘trained’ by the computed TECs from reference stations’ duel-frequency GPS data until the required accuracy was achieved. The experimental data are collected from 6 e-GPS stations of Tainan city government on January 3 to January 5, 2008. The input values for the ANN includ the geographical location of the ionosphere pierce point (IPP) and solar activity (sunspot number). The output value are those IPPs’ vertical total electron content (VTEC). Different times range and data types (IPPs’ or raster data) for the impact of the ANN are tested. And then compared to Klobuchar model and global ionopheric model, according to the correct rate and the ΔTEC statistic table decide the effectiveness of ANN. According to the test results, the regional ionopheric model constructed by ANN can corrected 80% of the ionospheric errors, the standard deviation of ΔTEC is less than ±3TECU.

With the aim for higher economical efficiency, considerable and radical changes have occurred in the worldwide electricity sector since the beginning of 1980s. By that time, the electricity sector has been controlled by the state-owned vertically integrated monopolies which manage and control all generation, transmission, distribution and retail activities and the consumers buy electricity with a price set by these monopolies in that system. After the liberalization and restructuring of the electricity power sector, separation and privatization of these activities have been widely seen. The main purpose is to ensure competition in the market where suppliers and consumers compete with each other to sell or buy electricity from the market and the consumers buy the electricity with a price which is based on competition and determined according to sell and purchase bids given by producers and customers rather than a price set by the government. Due to increasing competition in the electricity market, accurate electricity price forecasts have become a very vital need for all market participants. Accurate forecast of electricity price can help suppliers to derive their bidding strategy and optimally design their bilateral agreements in order to maximize their profits and hedge against risks. Consumers need accurate price forecasts for deriving their electricity usage and bidding strategy for minimizing their utilization costs. This thesis presents the determination of system day ahead price (SGOF) at the day ahead market and system marginal price (SMF) at the balancing power market in detail and develops artificial neural network models together with multiple linear regression models to forecast these electricity prices in Turkish electricity market. Also the methods used for price forecasting in the literature are discussed and the comparisons between these methods are presented. A series of historical data from Turkish electricity market is used to understand the characteristics of the market and the necessary input factors which influence the electricity price is determined for creating ANN models for price forecasting in this market. Since the factors influencing SGOF and SMF are different, different ANN models are developed for forecasting these prices. For SGOF forecasting, historical price and load values are enough for accurate forecasting, however, for SMF forecasting the net instruction volume occurred due to real time system imbalances is needed in order to increase the forecasting accuracy.

The present work focuses on the system identification method of aerodynamic parameter estimation which is used to calculate the stability and control derivatives required for aircraft flight mechanics. A new rotorcraft parameter estimation technique is proposed which uses a type of artificial neural network (ANN) called radial basis function network (RBFN). Rotorcraft parameter estimation using ANN is an unexplored research topic and the earlier works in this area have used the output error, equation error and filter error methods which are conventional parameter estimation methods. However, the conventional methods require an accurate non-linear rotorcraft simulation model which is not required by the ANN based method. The application of RBFN overcomes the drawbacks of multilayer perceptron (MLP) based delta method of parameter estimation and gives satisfactory results at either end of the ordered set of estimates. This makes the RBFN based delta method for parameter estimation suitable for rotorcraft studies, as both transition and high speed flight regime characteristics can be studied. The RBFN based delta method for parameter estimation is used for computation of aerodynamic parameters from both simulated and real time flight data. The simulated data is generated from an 8-DoF non-linear simulation model based on the Level-1 criteria of rotorcraft simulation modeling. The generated simulated data is used for computation of the quasi-steady and the time-variant stability and control parameters for different flight conditions using the RBFN based delta method. The performance of RBFN based delta method is also analyzed in the presence of state and measurement noise as well as outliers. The established methodology is then applied to compute parameters directly from real time flight test data for a BO 105 S123 helicopter obtained from DLR (German Aerospace Center). The parameters identified using the RBFN based delta method are compared with the identified values for the BO 105 helicopter from published literature which have used conventional parameter estimation techniques for parameter estimation using a 6-DoF and a 9-DoF rotorcraft simulation model. Finally, the estimated parameters are verified from the flight data generated by a frequency sweep pilot control input for assessing the predictive capability of the RBFN based delta method. Since the approach directly computes the parameters from flight data, it can be used for a reliable description of the higher frequency range, which is needed for high bandwidth flight control and in-flight simulation.