Yes / The advent of smart meter technology has enabled periodic monitoring of consumer energy consumption. Hence, short term energy forecasting is gaining more importance than conventional load forecasting. An Accurate forecasting of energy consumption is indispensable for the proper functioning of a virtual power plant (VPP). This paper focuses on short term energy forecasting in a VPP. The factors that influence energy forecasting in a VPP are identified and an artificial neural network based energy forecasting model is built. The model is tested on Sydney/ New South Wales (NSW) electricity grid. It considers the historical weather data and holidays in Sydney/ NSW and forecasts the energy consumption pattern with sufficient accuracy.

Yes / Real time traffic flow forecasting is a necessary requirement for traffic management in order to be able to evaluate the effects of different available strategies or policies. This paper focuses on short-term traffic flow forecasting by taking into consideration both spatial (road links) and temporal (lag or past traffic flow values) information. We propose a Layered Ensemble Model (LEM) which combines Artificial Neural Networks and Graded Possibilistic Clustering obtaining an accurate forecast of the traffic flow rates with outlier detection. Experimentation has been carried out on two different data sets. The former was obtained from real UK motorway and the later was obtained from simulated traffic flow on a street network in Genoa (Italy). The proposed LEM model for short-term traffic forecasting provides promising results and given the ability for outlier detection, accuracy, robustness of the proposed approach, it can be fruitful integrated in traffic flow management systems.

<p>Kreirana je specijalizovana baza podataka o sadržaju vitamina D koja sadrži 981 analitički podatak prikupljen iz evropskih baza podataka, od čega je 658 (67%) izraženo u formi ukupnog vitamina D. Podaci o svim vitamerima pronađeni su za meso, obogaćene namirnice/formulacije i za ribu, dok su podaci o D3 pronađeni za ribu, meso i mlečne proizvode. Podaci o sadržaju vitamina D, iz srpske baze podataka o sastavu namirnica (BPSN), ažurirani su za ukupno 541 namirnicu, jelo i dijetetski suplement.<br />Da bi se upotpunio kvalitet podataka o vitaminu D u srpskoj BPSN, određen je sadržaj vitamina D u svežim konzumnim koko&scaron;ijim jajima proizvedenim na teritoriji Srbije. Analizirana su jaja iz intenzivne proizvodnje i iz malih domaćinstava. Analize su izvedene u laboratoriji Danskog Tehničkog Univerziteta (DTU) standardizovanom HPLC metodom. Sadržaj vitamina D u jajima iz intenzivne proizvodnje iznosio je 5,78 &mu;g/100g, a u jajima iz slobodnog uzgoja 2,99 &mu;g/100 g. Izračunati prosečni sadržaj vitamina D u svežim jajima iznosio je 4,39 &mu;g/100 g te je ovaj podatak unet u srpsku BPSN.<br />Uobičajeni unos vitamina D analiziran je programom SPADE u populaciji koju su činili ispitanici iz četiri regiona Srbije, ukupno 605 odraslih, od toga 54% žena. Ustanovljeno je da uobičajeni prosečni unos vitamina D iznosi 4&plusmn;1,4 &mu;g/dan, kod mu&scaron;karaca 4,3&plusmn;1,5 &mu;g/dan, a kod žena 3,7&plusmn;1,2 &mu;g/dan, &scaron;to je znatno niže od preporučenih vrednosti od 10 &mu;g/dan za procenjene prosečne potrebe (Estimated Average Requirement-EAR) i 15 &mu;g/dan za adekvatni unos (Adequate Intake &ndash;AI). Čak 95% srpske populacije ne dostiže EAR vrednosti.<br />Analiza ishrane srpske populacije pokazala je da su glavni nutritivni izvori vitamina D jaja, riba, meso i mlečni proizvodi. Konzumacija obogaćenih namirnica vitaminom D (obogaćenih i biljnih mleka, kakao praha, obogaćenih sokova, margarina i instant žitarica) identifikovana je kod trideset petoro ispitanika. Prateći kriterijume za odabir adekvatnih namirnica za obogaćivanje, a za potrebe dizajniranja prediktivnog modela, odabrano je 70 namirnica koje su sortirane u sedam karakterističnih grupa: beli hleb, mleko, jogurt, sir, pavlaka, jaja i paradajz pire.<br />Prediktivni model obogaćivanja namirnica baziran je na matematičkoj formuli kojom se izračunava količina vitamina D (fc) koju treba dodati određenoj namirnici, odnosno grupi namirnica. Izračunata količina zavisi od tri faktora:<br />- prosečne konzumacije date namirnice, ili grupe, u gramima na n-tom percentilu populacije,<br />- njenog (njihovog) procentualnog udela u dnevnom energetskom unosu,<br />- unosa vitamina D (u &mu;g/dan) na n-tom percentilu.<br />Odabrano je sedam scenarija koji su simulirani da bi se validirala efektivnost &bdquo;dodavanja&ldquo; vitamina D radi dostizanja preporučenih nutritivnih vrednosti. U optimalnom scenariju, AI je dostignut na 65. percentilu populacije, a unos vitamina D na 95. percentilu populacije bio je ispod 25 &mu;g/dan. U maksimalnom scenariju, 50% populacije bilo je između AI i gornjeg tolerisanog nivoa nutritivnog unosa (Upper Tolerable Intake Level-UL), pri čemu niko nije dostigao UL vrednosti. Na ovaj način definisane su optimalne i maksimalne količine vitamina D koje se mogu dodati odabranim namirnicama da bi se zadovoljile potrebe, odnosno korigovao unos vitamina D kod srpske populacije.</p> / <p>A specialized database on the content of vitamin D was created with 981 analytical data on vitamin D content obtained from European databases, of which 658 (67%) were expressed as total vitamin D. The data (for all vitamins) were mainly found for meat, enriched foods/formulations and fish, while D3 data was identified for fish, meat and dairy products. Updating data on vitamin D content in Serbian food composition database (FCDB) was done in 541 foods, dishes and dietary supplements. To enhance the quality of data in Serbian FCDB, content of vitamin D in fresh eggs from the farm and domestic production on the territory of Serbia has been determined. Analysis was performed in Danish Technical University-DTU, Denmark, using standardized HPLC method. Eggs from the farm contained 5.78 &mu;g vitamin D/100 g, while domestic eggs were 2.99 &mu;g vitamin D/100 g, and the average vitamin D content in fresh eggs - 4.39 &mu;g/100 g which value was inserted into Serbian FCDB. The usual dietary intake of vitamin D was analyzed using the SPADE program in the survey covering 605 adult respondents from four regions of Serbia, of which 54% were women. The average intake of vitamin D was found to be 4&plusmn;1.4&mu;g/day, which is 4.3&plusmn;1.5 &mu;g/day for men and 3.7&plusmn;1.2 &mu;g/day for women, and is significantly lower than the recommended Estimated Average Requirement (EAR) values (10 &mu;g/day) and Adequate Intake (AI) values (15 &mu;g/day). As many as 95% of Serbian population are not reaching the EAR values. Nutritional analysis of Serbian diet has shown that the main sources of vitamin D are eggs, fish, meat and dairy products. Consumption of vitamin D-fortified foods (fortified and plant milk, cocoa powder, fortified juices, margarine, and instant cereals) was identified in 35 subjects. Following the criteria for selecting adequate foods for fortification, for the needs of designing the model, 70 foods were selected that were sorted into 7 characteristic food groups: white bread, milk, yoghurt, cheese, sour cream, eggs and tomato puree.<br />The prediction model of food fortification is based on a mathematical formula that calculates the amount of vitamin D (fc) to be added to a particular food group in accordance with:<br />- the amount of consumption of that food vector and<br />- the percentage factor in the total energy intake of the considered foods (food vectors) in the observed population,<br />- the intake of vitamin D on n-th percentile.<br />Seven scenarios were simulated to validate the effect of addition of vitamin D toward reaching the given reference values. In the optimal scenario, AI was reached at the 65th percentile of the population, and vitamin D intake at 95th percentile was below 25 &mu;g/day. In the maximum scenario, 50% of the population was between AI and Upper Tolerable Intake Level (UL), while none has reached UL values. This defines the ranges of optimal and maximum values of vitamin D that, by being added to the chosen food-vectors, can help in reaching vitamin D requirements of Serbian population.</p>

As crescentes preocupações com questões ambientais têm despertado interesse global pelo uso de combustíveis alternativos, e o uso da biomassa vegetal surge como uma alternativa viável para a geração de biocombustíveis. Diferentes materiais orgânicos têm sido utilizados, e dentre eles destaca-se o sorgo biomassa (Sorghum bicolor L. Moench). A seleção genômica apresenta grande potencial e pode, em médio prazo, reestruturar os programas de melhoramento de plantas, promovendo maiores ganhos genéticos quando comparada a outros métodos, além de reduzir significativamente o tempo necessário para o desenvolvimento de novas cultivares, através da seleção precoce. Este trabalho teve como objetivo avaliar modelos de seleção genômica e aplicá-los para a predição dos valores genéticos de indivíduos do painel de sorgo biomassa da Embrapa/Milho e Sorgo. Tal painel inclui materiais do banco de germoplasma e materiais utilizados em programas de melhoramento de sorgo dessa instituição, bem como coleções núcleo do CIRAD e ICRISAT, sendo, portanto, subdividido em dois sub-painéis. As 100 linhagens do sub-painel 1 foram avaliadas fenotipicamente por dois anos (2011 e 2012) e as 100 linhagens do sub-painel 2 por um ano (2011), ambas no município de Sete Lagoas-MG, para as seguintes características fenotípicas: tempo até o florescimento, altura de plantas, produção de massa verde e massa seca, proporções de fibra ácida e neutra, celulose, hemicelulose e lignina. Posteriormente, as 200 linhagens integrantes do painel foram genotipadas através da técnica de genotipagem por sequenciamento. A partir desses dados genotípicos e fenotípicos, os modelos de seleção genômica Bayes A, Bayes B, Bayes C&pi;, Bayes Lasso, Bayes Ridge Regression e Random Regression BLUP (RRBLUP) foram ajustados e comparados. As capacidades preditivas obtidas foram elevadas e pouco variaram entre os diversos modelos, variando de 0,61 para o caráter florescimento a 0,85 para a proporção de fibra ácida, quando o modelo RRBLUP foi empregado na análise conjunta dos dois sub-painéis. Por outro lado, a predição cruzada entre sub-painéis resultou em capacidades preditivas substancialmente menores, nunca superiores a 0,66 e em alguns cenários virtualmente iguais a zero, além de apresentar maiores variações entre os modelos ajustados. Simulações do uso de subconjuntos dos marcadores moleculares são apresentadas e indicam possibilidades de obtenção de capacidades preditivas mais elevadas. Análises de enriquecimento funcional realizadas a partir dos efeitos preditos dos marcadores sugeriram associações interessantes, as quais devem ser investigadas com maiores detalhes em estudos futuros, com potencial de elucidação da arquitetura genética dos caracteres quantitativos. / Increased concerns about environmental issues have aroused global interest in the use of alternative fuels, and the use of plant biomass emerges as a viable alternative for the generation of biofuels. Different organic materials have been used, including high biomass sorghum (Sorghum bicolor L. Moench). Genomic selection has great potential and could, in the medium term, restructure plant breeding programs, promoting greater genetic gains when compared to other methods and significantly reducing the time required for the development of new cultivars through early selection. This work aimed at evaluating models of genomic selection and applying them to the prediction of breeding values for a panel of high biomass sorghum genotypes of Embrapa / Milho e Sorgo. This panel includes materials from the gene bank and materials used in sorghum breeding programs of this institution, as well as core collections from CIRAD and ICRISAT, and is therefore divided into two sub-panels. The 100 lines of sub-panel 1 were evaluated phenotypically for two years (2011 and 2012) and the 100 lines of sub-panel 2 for one year (2011), both in the city of Sete Lagoas, Minas Gerais, for the following phenotypic traits: days to flowering, plant height, fresh and dry matter yield and fiber, cellulose, hemicellulose and lignin proportions. Subsequently, the 200 lines were genotyped by via the genotyping by sequencing technique. From these genotypic and phenotypic data, genomic selection models Bayes A, Bayes B, Bayes C&pi;, Bayes Lasso, Bayes Ridge Regression and Random Regression BLUP (RRBLUP) were fitted and compared. The predictive capabilities obtained were high and varied little between the different models, ranging from 0.61 for days to flowering to 0.85 for acid fiber, when the RRBLUP model was used on the combined analysis of the two sub-panels. On the other hand, cross prediction between sub-panels resulted in substantially lower predictive capability, never above 0.66 and in some scenarios virtually equal to zero, with greater variations between the fitted models. Simulations of using subsets of molecular markers are presented and indicate possibilities of achieving higher predictive capabilities. Functional enrichment analyses performed with the marker predicted effects suggested interesting associations, which should be investigated in more detail in future studies, with potential for elucidating the genetic architecture of quantitative traits.

Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2008.

DIRECT USE OF PGV FOR ESTIMATING PEAK NONLINEAR OSCILLATOR DISPLACEMENTS K&Uuml / &Ccedil / &Uuml / KDOGAN, Bilge Recently established approximate methods for estimating the lateral deformation demands on structures are based on the prediction of nonlinear oscillator displacements (Sd,ie). In this study, a predictive model is proposed to estimate the inelastic spectral displacement as a function of peak ground velocity (PGV). Prior to the generation of the proposed model, nonlinear response history analysis is conducted on several building models of wide fundamental period range and hysteretic behavior to observe the performance of selected demands and the chosen ground-motion intensity measures (peak ground acceleration, PGA, peak ground velocity, PGV and elastic pseudo spectral acceleration at the fundamental period (PSa(T1)). Confined to the building models used and ground motion dataset, the correlation studies revealed the superiority of PGV with respect to the other intensity measures while identifying the variation in global deformation demands of structural systems (i.e., maximum roof and maximum interstory drift ratio). This rational is the deriving force for proposing the PGV based prediction model. The proposed model accounts for the variation of Sd,ie for bilinear hysteretic behavior under constant ductility (&micro / ) and normalized strength ratio (R) associated with postyield stiffness ratios of = 0% and = 5%. Confined to the limitations imposed by the ground-motion database, the predictive model can estimate Sd,ie by employing the PGV predictions obtained from the attenuation relationships. This way the influence of important seismological parameters can be incorporated to the variation of Sd,ie in a fairly rationale manner. Various case studies are presented to show the consistent estimations of Sd,ie by the proposed model using the PGV values obtained from recent ground motion prediction equations.

Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2008.

Thesis (M.Sc. (Computer Science))--North-West University, Potchefstroom Campus, 2008.

A variety of high-permittivity (high-k) and low-loss ceramic-thermoplastic composite materials as fused deposition modeling (FDM) feedstock, based on cyclo-olefin polymer (COP) embedded with sintered ceramic fillers, have been developed and investigated for direct digital manufacturing (DDM) of microwave components. The composites presented in this dissertation use a high-temperature sintering process up to 1500°C to further enhance the dielectric properties of the ceramic fillers. The electromagnetic (EM) properties of these newly developed FDM composites were characterized up to the Ku-band by using the cavity perturbation technique. Several models for prediction of the effective relative dielectric permittivity of composites based on the filler loading volume fraction have been evaluated, among which Hanai-Bruggeman and Maxwell models have shown the best accuracy with less than 2% and 5% discrepancies, respectively. The 30 vol. % COP-TiO2 FDM-ready composites with fillers sintered at 1200°C have exhibited a relative permittivity (εr) of 4.78 and a dielectric loss tangent (tan δd) lower than 0.0012 at 17 GHz. Meanwhile, the 30 vol. % COP-MgCaTiO2 composites with fillers sintered at 1200°C have exhibited a εr of 4.82 and a tan δd lower than 0.0018. The DDM approach combines FDM of the engineered EM composites and micro-dispensing for deposition of conductive traces to fabricate by 3D-printing edge-fed patch antennas operating at 17.2 GHz and 16.5 GHz. These antennas were demonstrated by employing a 25 vol. % COP-MgCaTiO2 composite FDM filament with the fillers sintered at 1100°C and a pure COP filament, which were both prepared and extruded following the process described in this dissertation. The low dielectric loss of the 25 vol. % COP-MgCaTiO2 composite material (tan δd lower than 0.0018) has been leveraged to achieve a peak realized gain of 6 dBi. Also, the high-permittivity (εr of 4.74), which corresponds to an index of refraction of 2.17, results in a patch area miniaturization of 50% when compared with an antenna designed and DPAM-printed over a Rogers RT/duroid® 5870 laminate core through micro-dispensing of CB028 silver paste. This reference antenna exhibited a measured peak realized gain of 6.27 dBi that is comparable. Also, two low-loss FDM-ready composite materials for DDM technologies are presented and characterized at V-band mm-wave frequencies. Pure COP thermoplastic exhibits a relative permittivity εr of 2.1 and a dielectric loss tangent tan δd below 0.0011 at 69 GHz, whereas 30 vol. % COP-MgCaTiO2 composites with fillers sintered at 1200°C exhibit a εr of 4.88 and a tan δd below 0.0070 at 66 GHz. To the best of my knowledge, these EM properties (combination of high-k and low-loss) are superior to other 3D-printable microwave materials reported by the scientific microwave community and are on par with materials developed for high-performance microwave laminates by RF/microwave industry as shown in Chapter 5 and Chapter 7 and summarized in Table 5.4 and Table 7.1. Meanwhile, the linear coefficient of thermal expansion (CTE) from -25°C to 100°C of the reinforced 30 vol. % COP-MgCaTiO2 composite with fillers sintered at 1200°C is 64.42 ppm/°C, which is about 20 ppm/°C lower when compared with pure ABS and 10 ppm/°C lower as compared to high-temperature polyetherimide (PEI) ULTEM™ 9085 resin from Stratasys, Ltd. The CTE at 20°C of the same composite material is 84.8 ppm/°C which is about 20 ppm/°C lower when compared with pure ABS that is widely used by the research community for 3D printed RF/microwave devices by FDM. The electromagnetic (EM) composites with tailored EM properties studied by this work have a great potential for enabling the next generation of high-performance 3D-printed RF/microwave devices and antennas operating at the Ku-band, K-band, and mm-wave frequencies.

El Covid-19 es una enfermedad nueva que ha puesto en una situación muy difícil al mundo entero. La situación del Perú ha sido de las más difíciles, llegando a sumar a abril del 2021 más de 140,000 muertes según SINADEF, llegando a un nuevo pico de 783 muertes promedio día a marzo del 2021. Ante esta alta demanda de servicios de salud, los profesionales del sector toman la responsabilidad de priorizar la atención de los pacientes con protocolos subjetivos. El Instituto Nacional de Salud (INS), lugar donde se desarrolló el presente estudio, tiene entre sus responsabilidades la prestación de servicios de salud para el control de enfermedades transmisibles y el desarrollo de investigación científica-tecnológica. Por esto ha implementado laboratorios para pruebas moleculares, protocolos de atención, secuenciado el genoma de la variante peruana, etc. En esa línea el INS, además de facilitarnos la información de las pruebas Covid19 (Mas de 7 millones de registros con factores de riesgo, síntomas, edad, género, etc.) y los fallecidos del SINADEF nos proveyeron de juicio experto a través de sus profesionales: epidemiólogos, informáticos biomédicos, investigadores, etc. Los modelos predictivos permiten analizar grandes volúmenes de datos y a partir de ellos descubrir patrones y determinar sus correlaciones. Con toda la información disponible y aplicando los procesos de Business Analytics, se desarrolló un modelo analítico para la priorización de pacientes Covid-19. El modelo podría también ser utilizado en la prelación de la vacunación y en diseño de políticas públicas de salud relacionadas con la pandemia / Covid-19 is a new disease that has put the whole world in a very difficult situation. The situation in Peru has been one of the most difficult, reaching more than 140,000 deaths in April 2021 according to SINADEF, reaching a new peak of 783 average deaths per day in March 2021. Faced with this high demand for health services, the Industry professionals take responsibility for prioritizing patient care with subjective protocols. The National Institute of Health (INS), the place where this study was carried out, has among its responsibilities the provision of health services for the control of communicable diseases and the development of scientific-technological research. For this reason, it has implemented laboratories for molecular tests, care protocols, sequenced the genome of the Peruvian variant, etc. Along these lines, the INS, in addition to providing us with information on the Covid19 tests (more than 7 million records with risk factors, symptoms, age, gender, etc.) and the deceased from SINADEF, provided us with expert judgment through their professionals: epidemiologists, biomedical computer scientists, researchers, etc. Predictive models allow you to analyze large volumes of data and from them discover patterns and determine their correlations. With all the information available, applying the Business Analytics processes, an analytical model was developed for the prioritization of Covid-19 patients. The model could also be used in the priority of vaccination and in the design of public health policies related to the pandemic. / Trabajo de investigación