Time filter

Source Type

Lobao J.A.,Polytechnic Institute of Guarda | Devezas T.,University of Beira Interior | Catalao J.P.S.,University of Beira Interior | Catalao J.P.S.,University of Lisbon
Energy | Year: 2014

Increasing energy needs are accompanied by environmental responsibilities, since nowadays electricity companies operate in a competitive and sustainable energy framework. In this context, any proposal for action on energy efficiency becomes important for consumers to minimize operational costs. In electrical installations, electricity consumption can be decreased by reducing losses in the cables, associated with the overall efficiency of the equipment, allowing a better use of the installed power. The losses must be analysed in conjunction with all loads that contribute to the currents in the sections of an electrical installation. When replacing equipment in output distribution boxes with more efficient ones, the current in those sections is reduced in association with the decrease in power losses. This decrease, often forgotten, is taken into account in this work for the economic analysis of efficiency and sustainable electrical equipment. This paper presents a new software application that compares and chooses the best investment in the acquisition of electrical equipment. Simulation results obtained with the new software application are provided and are then validated with experimental results from a real electrical installation. © 2013 Elsevier Ltd.

Miguel S.P.,University of Beira Interior | Ribeiro M.P.,University of Beira Interior | Ribeiro M.P.,Polytechnic Institute of Guarda | Brancal H.,University of Beira Interior | And 3 more authors.
Carbohydrate Polymers | Year: 2014

Healing enhancement and pain control are critical issues on wound management. So far, different wound dressings have been developed. Among them, hydrogels are the most applied. Herein, a thermoresponsive hydrogel was produced using chitosan (deacetylation degree 95%) and agarose. Hydrogel bactericidal activity, biocompatibility, morphology, porosity and wettability were characterized by confocal microscopy, MTS assay and SEM. The performance of the hydrogel in the wound healing process was evaluated through in vivo assays, during 21 days. The attained results revealed that hydrogel has a pore size (90-400 μm) compatible with cellular internalization and proliferation. A bactericidal activity was observed for hydrogels containing more than 188 μg/mL of chitosan. The improved healing and the lack of a reactive or a granulomatous inflammatory reaction in skin lesions treated with hydrogel demonstrate its suitability to be used in a near future as a wound dressing. © 2014 Elsevier Ltd.

Marta C.,Polytechnic Institute of Guarda | Marinho D.A.,University of Beira Interior | Barbosa T.M.,Polytechnic Institute of Bragança | Izquierdo M.,Health Science University | Marques M.C.,University of Beira Interior
International Journal of Sports Medicine | Year: 2013

The purpose of this study was to compare the effects of an 8-weeks training period of resistance training alone (GR), combined resistance and endurance training (GCON) and a control group (GC) on explosive strength and VO 2max in a large sample of prepubescent boys and girls. 125 healthy children (58 boys, 67 girls), aged 10-11 years old (10.8±0.4 years) were assigned into 2 training groups to train twice a week for 8 weeks: GR (19 boys, 22 girls), GCON (21 boys, 24 girls) and a control group (GC: 18 boys, 21 girls; no training program). A significant but medium-sized increase from pre- to the post-training in the vertical jump (Effect size=0.22, F=34.44, p<0.01) and VO2max (Effect size=0.19, F=32.89, p<0.01) was observed. A significant large increase in the 1 kg (Effect size=0.53, F=202.17, p<0.01) and 3 kg (Effect size=0.48, F=132.1, p<0.01) ball throwing, standing long jump (Effect size=0.53, F=72.93, p<0.01) and running speed (Effect size=0.45, F=122.21, p<0.01) was also observed. The training group (GR and GCON) and sex factors did not significantly influence the evolution of strength variables from pre- to the post-training. The VO2max increased significantly only in GCON. Concurrent training is equally effective on training-induced explosive strength, and more efficient than resistance training only for VO 2max, in prepubescent boys and girls. This should be taken into consideration in order to optimize strength training school-based programs. © Georg Thieme Verlag KG Stuttgart. New York.

Gaspar P.D.,University of Beira Interior | Carrilho Goncalves L.C.,University of Beira Interior | Pitarma R.A.,Polytechnic Institute of Guarda
Applied Thermal Engineering | Year: 2011

The vertical open refrigerated display cabinets suffer alterations of their thermal performance and energy efficiency due to variations of ambient air conditions. The air curtain provides an aerothermodynamics insulation effect that can be evaluated by the thermal entrainment factor calculation as an engineering approximation or by the calculus of all sensible and latent thermal loads. This study presents the variation of heat transfer rate and thermal entrainment factor obtained through experimental tests carried out for different ambient air conditions, varying air temperature, relative humidity, velocity and its direction relatively to the display cabinet frontal opening. The thermal entrainment factor are analysed and compared with the total sensible and latent heats results for the experimental tests. From an engineering point of view, it is concluded that thermal entrainment factor cannot be used indiscriminately, although its use is suitable to design better cabinet under the same climate class condition. © 2010 Elsevier Ltd. All rights reserved.

Amado L.,Polytechnic Institute of Guarda | Albuquerque A.,University of Beira Interior | Espirito Santo A.,University of Beira Interior
Ecological Engineering | Year: 2012

This research work evaluates the impact of stormwater infiltration on the removal of organics, solids, nitrogen and phosphorus in a LECA-based horizontal subsurface flow constructed wetland. Stormwater runoff for the period between January 2008 and June 2010 (30 months) was estimated for the drainage basin of the Vila Fernando wastewater treatment plant (constructed wetland system) and it was observed proportionality between the increase in surface runoff and the infiltration flow rate that reached the plant. The average stormwater infiltration rate that reached the plant was 67.4m 3d -1 (77% of the influent flow rate into the plant). The 30 month monitoring campaign set up at one of the reed beds showed that stormwater infiltration led to a high variation of the hydraulic loading rate (HLR) throughout the bed, which affected its performance in the removal of organic matter, suspended solids and nitrogen. For a HLR below 20cmd -1 the removal efficiencies for BOD 5, COD and TSS doubled. The removal of nitrogen was too low and was associated with the weak development of nitrifying biofilm and the poor growth of reeds. However, it was observed a good relationship between the applied and removed loads for most of the parameters, which seems to indicate that the organic, solid, nitrogen and phosphorus loading rates influenced the respective removal rates. The use of a bed material with high specific surface area, such as LECA, can only improve the treatment capacity of horizontal subsurface flow systems if the organic and solid loading rates, as well as the HLR, can be effectively controlled. © 2011 Elsevier B.V.

Malonek H.R.,University of Aveiro | Tomaz G.,Polytechnic Institute of Guarda
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2011

Recently the creation matrix, intimately related to the Pascal matrix and its generalizations, has been used to develop matrix representations of special polynomials, in particular Appell polynomials. In this paper we describe a matrix approach to polynomials in several hypercomplex variables based on special block matrices whose structures simulate the creation matrix and the Pascal matrix. We apply the approach to hypercomplex Laguerre polynomials, although it can be used for other Appell sequences, too. © 2011 Springer-Verlag.

Tomaz G.,Polytechnic Institute of Guarda | Malonek H.R.,University of Aveiro
AIP Conference Proceedings | Year: 2010

Recently the Pascal matrix and its generalizations have been used to construct matrix representations of special polynomials in one variable like Bernoulli, Euler, Genocchi, Hermite, etc. In this paper we describe a unified approach via a block creation matrix to polynomials in several hypercomplex variables. © 2010 American Institute of Physics.

Caseiro L.M.A.,University of Coimbra | Mendes A.M.S.,University of Coimbra | Alcaso A.N.,Polytechnic Institute of Guarda
IECON Proceedings (Industrial Electronics Conference) | Year: 2012

Some of the main concerns in the use of power converters are their reliability and the ability to detect and maintain operation when in fault. Many techniques have been proposed to identify and respond to faults in Neutral-Point-Clamped converters. Most of these methods, however, require additional hardware and/or changes in the topology. In this paper, a NPC rectifier is investigated, including the impacts of each possible semiconductor fault in the operation. Methods for fault diagnosis and fault tolerant operation are presented. The presented diagnostic method does not require additional hardware and is not affected by the load. Open-circuit and short-circuit faults are detected in both IGBTs and diodes. Fault tolerant operation with different levels of action is presented. Experimental results are presented to demonstrate the effectiveness of the methods. © 2012 IEEE.

Lopes N.,University of Coimbra | Lopes N.,Polytechnic Institute of Guarda | Ribeiro B.,University of Coimbra
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) | Year: 2010

Non-Negative Matrix Factorization (NMF) algorithms decompose a matrix, containing only non-negative coefficients, into the product of two matrices, usually with reduced ranks. The resulting matrices are constrained to have only non-negative coefficients. NMF can be used to reduce the number of characteristics in a dataset, while preserving the relevant information that allows for the reconstruction of the original data. Since negative coefficients are not allowed, the original data is reconstructed through additive combinations of the parts-based factorized matrix representation. A Graphics Processing Unit (GPU) implementation of the NMF algorithms, using both the multiplicative and the additive (gradient descent) update rules is presented for the Euclidean distance as well as for the divergence cost function. The performance results on an image database demonstrate extremely high speedups, making the GPU implementations excel by far the CPU implementations. © 2010 Springer-Verlag Berlin Heidelberg.

Lopes N.,University of Coimbra | Lopes N.,Polytechnic Institute of Guarda | Ribeiro B.,University of Coimbra
International Journal of Neural Systems | Year: 2011

The Graphics Processing Unit (GPU) originally designed for rendering graphics and which is difficult to program for other tasks, has since evolved into a device suitable for general-purpose computations. As a result graphics hardware has become progressively more attractive yielding unprecedented performance at a relatively low cost. Thus, it is the ideal candidate to accelerate a wide variety of data parallel tasks in many fields such as in Machine Learning (ML). As problems become more and more demanding, parallel implementations of learning algorithms are crucial for a useful application. In particular, the implementation of Neural Networks (NNs) in GPUs can significantly reduce the long training times during the learning process. In this paper we present a GPU parallel implementation of the Back-Propagation (BP) and Multiple Back-Propagation (MBP) algorithms, and describe the GPU kernels needed for this task. The results obtained on well-known benchmarks show faster training times and improved performances as compared to the implementation in traditional hardware, due to maximized floating-point throughput and memory bandwidth. Moreover, a preliminary GPU based Autonomous Training System (ATS) is developed which aims at automatically finding high-quality NNs-based solutions for a given problem. © 2011 World Scientific Publishing Company.

Loading Polytechnic Institute of Guarda collaborators
Loading Polytechnic Institute of Guarda collaborators