Published Article / Early indication of possible drought can help in developing suitable drought mitigation strategies and measures in advance. Therefore, drought forecasting plays an important role in the planning and management of water resource in such circumstances. In this study, a non-linear streamflow forecasting model was developed using Artificial Neural Network (ANN) modeling technique at the Melka Sedi stream gauging station, Ethiopia, with adequate lead times. The available data was divided into two independent sets using a split sampling tool of the neural network software. The first data set was used for training and the second data set, which is normally about one fourth of the total available data, was used for testing the model. A one year data was set aside for validating the ANN model. The streamflow predicted using the model on weekly time step compared favorably with the measured streamflow data (R2 = 75%) during the validation period. Application of the model in assessing appropriate agricultural water management strategies for a large-scale irrigation scheme in the Awash River Basin, Ethiopia, has already been considered for publication in a referred journal.

Over recent years the delivery of nanosized drug particles has shown potential in improving bioavailability. Drug nanosizing is achieved by 'top-down' and by 'bottom-up' approaches. Owing to limitations associated with the top-down techniques, such as high energy input, electrostatic effects, broad particle size distributions and contamination issues, great interest has been directed to alternative bottom up technologies. In this study, the hypothesis that microreactors can be used as a simple and cost-effective technique to generate organic nanosized products is tested using three steroids (hydrocortisone, prednisolone and budesonide). Arrested antisolvent nanoprecipitation using ethanol (solvent) and water (antisolvent) was conducted within the microreactors. To enable experimental design for the microreactor studies, solubility profiles in different ethanol-water combinations at 25 °C were explored. All three drugs' solubility increased with increasing ethanol concentration showing maxima at 80-90 % v/v ethanol-water mixtures. Because of the complex multivariate microfluidic process, artificial neural network modelling was then employed to identify the dominant relationships between the variables affecting nanoprecipitation (as inputs) and the drug particle size (as output). The antisolvent flow rate was found to have the major role in directing drug particle size. Based on these successful findings, the potential of preparing pharmaceutical nanosuspensions using microfluidic reactors was researched. A hydrocortisone (HC) nanosuspension (NS) was prepared by introducing the generated drug particles into an aqueous solution of stabilizers stirred at high speed with a propeller mixer. A tangential flow filtration system was then used to concentrate the prepared NS. Results showed that a stable narrow sized HC NS of amorphous spherical particles 500 ± 64 nm diameter and zeta potential -18 ± 2.84 mV could be produced. The ocular bioavailability of a microfluidic precipitated HC NS (300 nm) was assessed and compared to a similar sized, milled HC NS and HC solution as a control. The precipitated and the milled NS achieved comparable AUC0-9h of 28.06 ± 4.08 and 30.95 ± 2.2, respectively, significantly (P < 0.01) higher than HC solution (15.86 ± 2.7). These results illustrate the opportunity to design sustained release ophthalmic formulations. Going nano via microfluidic precipitation was also exploited to tailor budesonide (BD) NS for pulmonary administration. The in vitro aerosolization by nebulization of a BD NS was studied in comparison with a commercial BD microsuspension. Overall, the fine particle fraction generated from BD NS (56.88 ± 3.37) was significantly (P < 0.05) higher than the marketed BD (38.04 ± 7.81). The mean mass aerodynamic diameter of BD NS aerosol (3.9 ± 0.48 μm) was significantly smaller (P < 0.05) than the microsuspension (6.2 ± 1.09 μm) indicating improved performance for BD NS. In conclusion, findings of this study support the hypothesis of using microfluidic nanoprecipitation as a promising and economical technique of drug nanosizing.

Nowadays islanding has become a big issue with the increasing use of distributed generators in power system. In order to effectively detect islanding after DG disconnects from main source, author first studied two passive islanding methods in this thesis: THD&VU method and wavelet-transform method. Compared with other passive methods, each of them has small non-detection zone, but both of them are based on the threshold limit, which is very hard to set. What’s more, when these two methods were applied to practical signals distorted with noise, they performed worse than anticipated. Thus, a new composite intelligent based method is presented in this thesis to solve the drawbacks above. The proposed method first uses wavelet-transform to detect the occurrence of events (including islanding and non-islanding) due to its sensitivity of sudden change. Then this approach utilizes artificial neural network (ANN) to classify islanding and non-islanding events. In this process, three features based on THD&VU are extracted as the input of ANN classifier. The performance of proposed method was tested on two typical distribution networks. The obtained results of two cases indicated the developed method can effectively detect islanding with low misclassification.

Este trabalho propõe o desenvolvimento de uma metodologia para o cálculo do volume de tráfego durante o período de inicialização de uma rede Profinet. O tráfego de dados é um dos indicadores de desempenho criados para garantir a qualidade dos protocolos baseados em Real Time Ethernet (RTE). Neste contexto, buscou-se na literatura uma forma de classificar o tráfego de acordo com a sua magnitude e mensurar seu efeito na comunicação. Dados provenientes de redes criadas em laboratório foram coletados e aplicados a uma Rede Neural Artificial visando generalizar o conhecimento adquirido. O uso dado a RNA foi de estimação da função de interesse. Os resultados obtidos após o processamento dos dados reais são considerados satisfatórios e condizentes às expectativas dessa dissertação, já que se buscou, por razões inerentes ao problema estudado, um erro relativo inferior 3%. Conclui-se que a metodologia apresentada é factível e aplicável ao meio industrial, podendo ser parte de uma ferramenta mais completa, como os analisadores de redes Profinet. / This paper suggests the development of a methodology to calculate the traffic volume during the starting period of a Profinet network. The data traffic is one of the development indicators created to guarantee the protocols quality based on Real Time Ethernet (RTE). In this context, a way of classifying the traffic according to its magnitude and of measuring its effect in the communication was searched in the literature. Data deriving from networks created in laboratory were collected and applied into an Artificial Neural Network aiming to generalize the acquired knowledge. The ANN was used to estimate the function of interest. The results obtained after the real data processing are considered satisfactory and suitable to the expectations of this dissertation where the relative error inferior to 3%, for reasons intrinsic to the studied problem, was searched. It is concluded that the methodology presented is feasible and applicable in the industrial field, where it can be part of a more complete tool, as the Profinet network analyzers.

Deterministiska turbaserad tvåspelarspel är ett område som används inom AI forskning för att jämföra AI system. Detta arbete fokuserar på att jämföra teknikerna artificiell neuralt nätverk och alpha-beta pruning i spelet othello. Arbetet undersökte hur dessa tekniker presterar i relation till beräkningstiden. Othello positionerna representeras i en 8 x 8 matris som teknikerna använder för att hitta det optimala draget. Systemen värderades enligt en definierat metod som använder ett befintlig AI system för othello Edax. De testades på 154 othello partier med 77 stycken förbestämda startpositioner. Nätverket tränades med inlärningsdata som bestod av drag från professionella othello matcher och från Edax. Resultatet visade att ABP systemen värderades linjärt mot exponentiell beräkningstid medans ANN systemen värderades konstant mot linjär beräkningstid. Resultatet av ANN systemen tyder på att inlärningsdatan är bristande. Framtida arbete bör använda mer och bättre inlärningsdata.

Det här experimentet gick ut på att testa tre olika inlärningsstrategier emot varandra i en spelmiljö. De tre inlärningsstrategier som testades var en exempelbaserad strategi, en belöningsbaserad strategi och en strategi som kombinerade dessa två algoritmer. Kombinationen bestod av att först träna upp ett nätverk med den exempelbaserade strategin för att sedan använda det nätverket som en utgångspunkt för den belöningsbaserade strategin. Dessa strategier testades sedan i en spelmiljö. Resultatet som de olika strategierna producerade var svårtolkade. På grund av att alla inlärningsstrategierna inte förbättrade sig märkvärdigt under deras träningstid har det gjort att resultaten från att ändra på variablerna hos de olika strategierna inte visat på några märkvärda skillnader i resultat. Skillnader i resultat vid jämförelser av de olika strategierna har observerats men det går inte att säkerställa att det är inlärningsstrategierna som är anledningen till skillnad i resultat.

Este trabalho tem como objetivo analisar o conceito de autoria no contexto da obra de George Eliot. Para realizá-lo, definiram-se dois aspectos relevantes da atividade do autor. O primeiro deles é sua existência empírica, situada dentro de determinadas práticas, como apontado por Martha Woodmansee, Michel Foucault, Marisa Lajolo e Regina Zilberman. O segundo, seu funcionamento intratextual, como instância discursiva, destilado das teorias enunciativas de Emile Benveniste e das proposições teóricas de Wayne Booth, Umberto Eco e Wolfgang Iser. A partir dessas elaborações, foram desenvolvidas as análises, de um lado, da trajetória de Mary Ann Evans (George Eliot) e seu papel como escritora do final do século XIX, e, de outro, do romance The Mill on the Floss, obra de 1860, na qual se percebe o autor George Eliot em funcionamento. Pode-se notar, através da reconstrução da vida da autora, sua reflexão própria sobre o significado da prática da autoria como missão social. É também notável, através de seu romance, a atuação de uma figura autoral que organiza o texto e encaminha a interpretação de seu leitor para determinadas direções. / This work’s objective is to analyze the concept of authorship in the context of George Eliot’s production. In order to do so, two relevant aspects of the author’s activity were defined. The first one is its empirical existence, located within certain practices, as pointed by Martha Woodmansee, Michel Foucault, Marisa Lajolo and Regina Zilberman. The second one, its intratextual operation, as a discoursive instance, distilled from Emile Benveniste’s enunciative theories and from the theoretical propositions of Wayne Booth, Umberto Eco and Wolfgang Iser. These elaborations allowed the development of two analyses: on one hand, of Mary Ann Evans’ (George Eliot’s) trajectory and her role as a late nineteenth century writer, and, on the other, of the novel The Mill on the Floss (1860), in which George Eliot’s authorship is perceived at work. It is noticeable, through the reconstruction of the author’s life, her own reflection on the meaning of authorial practice as a social mission. It is also remarkable, through her novel, the performance o an author figure which organizes the text and directs its reader’s interpretations to determined directions.

Menée en partenariat avec Thales Alenia Space et le Centre National d’Etudes Spatiales (CNES), cette thèse s’inscrit dans un contexte international très actif sur une nouvelle technologie de systèmes antennaires : les réseaux réflecteurs (RA). Combinant les atouts des réflecteurs classiques et ceux des réseaux, les RA pourraient remplacer, à terme, les réflecteurs formés utilisés aujourd’hui. Ils sont constitués d’une source primaire placée en regard d’un réseau de cellules contrôlant les propriétés du champ électromagnétique réfléchi. Si de nombreuses études portent déjà sur la caractérisation des cellules, une des problématiques consiste à les sélectionner judicieusement pour réaliser l’antenne finale : c’est l’étape de synthèse.Un tour d’horizon consacré aux différentes méthodes de synthèse a révélé toute la complexité d’obtenir rapidement de bonnes performances simultanément sur les polarisations directes et croisées, pour une large bande de fréquences et pour la réalisation de diagrammes formés. La cellule Phoenix est sélectionnée dans ce travail pour ses bonnes propriétés puisqu’elle fournit l’ensemble de la gamme de phase suivant un cycle continu de géométries.Parce qu’une des contraintes dans la conception de RA est de maintenir des variations de géométries continues entre deux cellules juxtaposées à la surface du RA, un outil de représentation sphérique permettant de classer l’ensemble des cellules d’étude est proposé. Il répertorie judicieusement toutes les cellules sur une surface continue, fermée et périodique.Une nouvelle étape est franchie avec la conception de modèles comportementaux à l’aide de réseaux de neurones (ANN). Ces modèles permettent une caractérisation électromagnétique très rapide des cellules en termes de phase et d’amplitude des coefficients directs et croisées de la matrice de réflexion.L’originalité de l’algorithme de synthèse proposé dans ce travail se base sur l’utilisation combinée de la représentation sphérique et de la caractérisation rapide par ANN. Un outil d’optimisation de type min / max est utilisé pour améliorer les performances globales du panneau RA. Il est ensuite appliqué à un cas concret dans le cadre d’une mission de télécommunication. / In collaboration with Thales Alenia Space and the French Space Agency (CNES), this PHD takes part in a very active international context on a new antenna: the reflectarrays (RA).Combining the advantages of conventional reflectors and those of networks, RA could eventually replace the currently used shaped reflectors. They consist of a primary source placed in front of a network of cells controlling the properties of the reflected electromagnetic field. Although many studies already focus on the characterization of cells, one of the issues is to carefully select them to achieve the final antenna: this is the synthesis step.An overview of different synthesis methods revealed the complexity to quickly obtain good performance simultaneously on the co- and cross-polarizations, for a wide frequency band and for the realization of shaped radiation pattern. The Phoenix cell is selected in this work for its good properties since it provides the entire phase range following a continuous cycle of geometries.Because one of the constraints in the design of RA is to maintain continuous geometry variations between two juxtaposed cells on the layout, a spherical representation tool made it possible to classify all the studied cells. It judiciously lists all the cells on a continuous, closed and periodic surface.A new step is reached with the design of behavioral models using Artificial Neural Networks (ANN). These models enable to a fast electromagnetic characterization of cells in terms of phase and amplitude of the direct and cross coefficients of the reflection matrix.The originality of the synthesis algorithm proposed in this work is the combined use of the spherical representation and a rapid cell characterization by ANN. A min / max optimization tool is used to improve the overall performance of the RA panel. It is then applied to a concrete case as part of a telecommunication mission.

This essay presents an ecocritical reading of water imagery in selected poems by Carol AnnDuffy, with focus on Duffy’s personified water-voices, how water illuminates history, andDuffy’s metaphor of language as water. After a consideration of the problematics of teachingpoetry in the EFL classroom, the essay concludes that Duffy’s poetry holds potential forstudents to develop environmental literacy, both in content and in form.

Three dimensional (3D) ground reaction force (GRF) are an essential component for gait analysis. Current methods for measuring 3D GRF involve using a stationary force plate embedded in the ground, which captures the forces as subjects walk across the platform. This approach has several limitations, a few being: it can only capture a few steps at a time, it is expensive to purchase and maintain, it can't reflect forces caused by natural uneven surfaces, etc. Previous research has attempted to develop wearable force sensors to overcome these problems; however, these endeavors have resulted in devices that are expensive, bulky, and fail to accurately measure forces when compared to static force plates. This thesis presents the implementation and validation of novel nanocomposite piezoresponsive foam (NCPF) sensors for measuring 3D GRF. Four NCPF sensors were embedded in a shoe sole at four locations: heel, arch, ball, and toe. The signals from each sensor were used in a functional data analysis (FDA) to develop a statistical model for estimating 3D GRF. The process of calibrating the sensors to model GRF was validated through a study where 9 subjects (4 females, 5 males) walked on a force-sensing treadmill for two minutes. Two approaches were used to model the GRF response. The first approach was based on functional decomposition of the data. Using a tenfold cross validation process a statistical model was developed for each subject with the ability to predict walking 3D GRF with less than 7% error. The second approach used machine learning to model 3D GRF. Using the same walking data for the statistical models, an artificial neural network (ANN) was used to create subject-specific models that could predict walking 3D GRF with less than 11% error. The predictive capabilities of ANN were tested using a pilot study where a single subject performed a calibration procedure by running at seven different speeds for thirty seconds each on the force-sensing treadmill. This calibration data was used to train a model, which was then used to estimate vertical GRF (VGRF) for three additional running trials at randomly selected speeds from within the calibration range. The ANN model was able to predict VGRF for three running speeds after calibration with less than 4% error. The use of NCPF sensors to estimate 3D GRF was shown to be a viable alternative to static force plates. It is recommended, in future work, that 3D GRF and subsequent sensor data be collected from a large sample of subjects to create a baseline of 3D GRF characteristics for a population that will enable a robust cross-subject model capable of performing real-time ground reaction force analysis across the general population, which will greatly benefit our understanding of human gait.