Use of uniform designs in combination with neural networks for viral infection process development

This work aimed to compare the predictive capacity of empirical models, based on the uniform design utilization combined to artificial neural networks with respect to classical factorial designs in bioprocess, using as example the rabies virus replication in BHK-21 cells. The viral infection process parameters under study were temperature (34°C, 37°C), multiplicity of infection (0.04, 0.07, 0.1), times of infection, and harvest (24, 48, 72 hours) and the monitored output parameter was viral production. A multilevel factorial experimental design was performed for the study of this system. Fractions of this experimental approach (18, 24, 30, 36 and 42 runs), defined according uniform designs, were used as alternative for modelling through artificial neural network and thereafter an output variable optimization was carried out by means of genetic algorithm methodology. Model prediction capacities for all uniform design approaches under study were better than that found for classical factorial design approach. It was demonstrated that uniform design in combination with artificial neural network could be an efficient experimental approach for modelling complex bioprocess like viral production. For the present study case, 67% of experimental resources were saved when compared to a classical factorial design approach. In the near future, this strategy could replace the established factorial designs used in the bioprocess development activities performed within biopharmaceutical organizations because of the improvements gained in the economics of experimentation that do not sacrifice the quality of decisions.

Individual and contextual factors related to dental caries in underprivileged Brazilian adolescents

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Multi-task control strategy for grid-tied inverters based on conservative power theory

Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP)

Projeto de controlador gain scheduling usando realimentação derivativa via LMI

Pós-graduação em Engenharia Elétrica - FEIS

Projeto de controladores robustos para acionamento de um motor de indução trifásico

Pós-graduação em Engenharia Elétrica - FEIS

A influência das variáveis do lixamento tubular na qualidade das peças de Eucalyptus grandis

In 2010, the Brazilian forest sector is represented by about 30,000 companies producing US$ 21 billion annually and account for approximately 5% of the gross domestic product (GDP) in the country. The sanding process is highly demanded in various stages of industrialization of the wood, when there is a need for a better quality surface finishing. The objective of this work was to analyze the influence of cutting speed and sandpaper granulometry on both the surface finishing of pieces of Eucalyptus grandis processed through tubular sanding and on the sanding efforts (force and power of sanding). Four cutting speeds were used (19.5, 22.7, 26 and 28.1 m/s), one advance speed (16 m/min) and three sets of sandpaper (80-100, 80-120 and 100-120) being one for chipping and another for finishing, respectively. A central data acquisition system was set up to capture the variables (cutting power, acoustic emission and vibration) in real time. The cutting force was obtained indirectly, through a frequency inverter. The roughness of the parts was measured by a roughness meter before and after sanding. The highest cutting speed used (28.1 m/s) consumed more power and generated more acoustic emission among the four speeds tested. Regarding the vibration, the lower cutting speed (19.5 m/ s) generated the highest vibration in the sander machine. It is concluded that the range of 100-120 sandpapers resulted in values of average roughness (Ra) lower than the other sets of sandpaper used, as it resulted in better surface finishing.

Estratégia de controle multifuncional para sistemas fotovoltaicos de geração de energia elétrica

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Microinversor integrado ao módulo fotovoltaico para sistemas conectados à rede elétrica utilizando controlador ressonante

This paper shows the modeling and control of a single-phase full-bridge inverter with high-frequency transformer that may be used as part of a two-stage converter with transformerless DC-DC side or as a single-stage converter (simple DC-AC converter) for grid-connected PV applications. The inverter is modeled in order to obtain a small-signal transfer function used to design the P+Resonant current controller. A highfrequency step-up transformer results in reduced voltage switches and better efficiency compared with converters in which the transformer is used on the DC-DC side. Simulations and experimental results with a 200 W prototype are shown. 1

Efeitos das correlações inter-átomos adsorvidos na densidade de estados do grafeno

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Proposta de equipamento de conexão à rede para a utilização da geração de energia solar em consumidores de pequeno e médio porte

Pós-graduação em Engenharia Mecânica - FEG

Modeling spatial distribution of cancer incidence in Cordoba, Argentina

Background: Cancer is the second leading cause of death in Argentina, and there is little knowledge about its incidence. The first study based on population-based cancer registry described spatial incidence and indicated that there existed at least county-level aggregation. The aim of the present work is to model the incidence patterns for the most incidence cancer in Córdoba Province, Argentina, using information from the Córdoba Cancer Registry by performing multilevel mixed model approach to deal with dependence and unobserved heterogeneity coming from the geo-reference cancer occurrence. Methods: Standardized incidence rates (world standard population) (SIR) by sex based on 5-year age groups were calculated for 109 districts nested on 26 counties for the most incidence cancers in Cordoba using 2004 database. A Poisson twolevel random effect model representing unobserved heterogeneity between first level-districts and second level-counties was fitted to assess the spatial distribution of the overall and site specific cancer incidence rates. Results: SIR cancer at Córdoba province shown an average of 263.53±138.34 and 200.45±98.30 for men and women, respectively. Considering the ratio site specific mean SIR to the total mean, breast cancer ratio was 0.25±0.19, prostate cancer ratio was 0.12±0.10 and lower values for lung and colon cancer for both sexes. The Poisson two-level random intercepts model fitted for SIR data distributed with overdispersion shown significant hierarchical structure for the cancer incidence distribution. Conclusions: a strong spatial-nested effect for the cancer incidence in Córdoba was observed and will help to begin the study of the factors associated with it.

Conversor inversor integrado três estados buck-boost para geração distribuída, com operação conectada ou ilhada

Pós-graduação em Engenharia Elétrica - FEIS