A novel forward gene selection algorithm for microarray data

This paper investigates the gene selection problem for microarray data with small samples and variant correlation. Most existing algorithms usually require expensive computational effort, especially under thousands of gene conditions. The main objective of this paper is to effectively select the most informative genes from microarray data, while making the computational expenses affordable. This is achieved by proposing a novel forward gene selection algorithm (FGSA). To overcome the small samples' problem, the augmented data technique is firstly employed to produce an augmented data set. Taking inspiration from other gene selection methods, the L2-norm penalty is then introduced into the recently proposed fast regression algorithm to achieve the group selection ability. Finally, by defining a proper regression context, the proposed method can be fast implemented in the software, which significantly reduces computational burden. Both computational complexity analysis and simulation results confirm the effectiveness of the proposed algorithm in comparison with other approaches

基于局部损伤混凝土模型的FRP-混凝土界面有限元分析研究

A meso-scale finite element model is presented for investigating the FRP-concrete bond behaviour under static and dynamic loadings. It adopts a local concrete damage model. A large number of single shear tests under static pull-off loading were modeled. It is demonstrated that the developed model can satisfactorily simulate the static debonding behaviour, in terms of the load-carrying capacity and the local bond-slip behaviour. A preliminary study of the effect of the dynamic loading rate on the debonding behaviour was also conducted by considering a dynamic increase factor for the concrete strength as a function of strain rate. It is shown that a higher loading rate leads to a higher load-carrying capacity, a longer effective bond length, and a larger damaged area of concrete in the single shear loading scenario.

Experimental study on PC hollow-core slab strengthened with bamboo plates

An experimental study on strengthening prestressed concrete (PC) hollow-core slabs was conducted. Nine PC hollow-core slabs were tested, including three unstrengthened reference slabs and six slabs strengthened with bamboo plates. The results show that compared with unreinforced slabs, the cracking loads of PC hollow-core slabs strengthened with bamboo plates increase by 5% to 96% (with an average of 41%), the loads at allowable deflection increase by 8% to 76% (with an average of 35%), and the ultimate loads increase by 83% to 184% (with an average of 123%), respectively. All the degrees of improvement in the crack load, allowable load and ultimate load increase with the increase in the thickness and width of the bamboo plates. With the increase in the loads, the strain distribution along the height of the strengthened slabs at the mid-span basically remains a plan-assumption. With the increase in the thickness and width of the bamboo plates, both the bamboo tensile strain on the tensile face and the concrete compressive strain on the compression face of the strengthened slabs decrease under the same load level.

FRP-to-concrete bond-slip behaviour under dynamic loading

Strengthening reinforced concrete (RC) structures by externally bonded FRP composites has been widely used for static loading and seismic retrofitting since 1990s. More recently many studies on strengthening concrete and masonry structures with externally bonded FRP for improved blast and impact resistance in protective engineering have also been conducted. The bond behaviour between the FRP and concrete plays a critical role in a strengthening system with externally bonded FRP. However, the understanding of how the bond between FRP and concrete performs under high strain rate is severely limited. Due to the dynamic characteristics of blast and impact loading, the bond behaviour between FRP and concrete under such loading is very different from that under static loading. This paper presents a study on the dynamic bond-slip behaviour based on both the numerical analysis and test results. A dynamic bond-slip model is proposed in this paper.

Strengthening timber beams with near surface mounted carbon fiber reinforced polymer rods

Many timber structures may require strengthening due to either decay and aging or an increase of load. This paper presents an experimental study in which eleven timber beams were tested, including three unstrengthened reference beams and eight beams strengthened with NSM CFRP bars. The test parameters include the position of NSM (tensile face or the bottom of the sides), the number of CFRP bars (1 or 2), and additional anchorage of NSM CFRP bars (steel wire U anchors or CFRP U strips). The test results show that the ultimate flexural strength of the timber beams were increased by 14%∼85% with an average of 47% due to NSM CFRP bar strengthening. Their deflection corresponding to the peak load was increased by 33% in average.

K&C局部损伤混凝土材料模型在精细有限元模拟中的应用

This paper presents a study on concrete fracture and the associated mesh sensitivity using the finite element (FE) method with a local concrete model in both tension (Mode I) and compression.To enable the incorporation of dynamic loading, the FE model is developed using a transient dynamic analysis code LS-DYNA Explicit.A series of investigations have been conducted on typical fracture scenarios to evaluate the model performances and calibration of relevant parameters.The K&C damage model was adopted because it is a comprehensive local concrete model which allows the user to change the crack band width, fracture energy and rate dependency of the material.Compressive localisation modelling in numerical modelling is also discussed in detail in relation to localisation.An impact test specimen is modelled.

Type Ia supernovae from exploding oxygen-neon white dwarfs

Context. The progenitor problem of Type Ia supernovae (SNe Ia) is still unsolved. Most of these events are thought to be explosions of carbon-oxygen (CO) white dwarfs (WDs), but for many of the explosion scenarios, particularly those involving the externally triggered detonation of a sub-Chandrasekhar mass WD (sub-M-Ch, WD), there is also a possibility of having an oxygen-neon (ONe) WD as progenitor.

Aims. We simulate detonations of ONe WDs and calculate synthetic observables from these models. The results are compared with detonations in CO WDs of similar mass and observational data of SNe Ia.

Methods. We perform hydrodynamic explosion simulations of detonations in initially hydrostatic ONe WDs for a range of masses below the Chandrasekhar mass (M-Ch), followed by detailed nucleosynthetic postprocessing with a 384-isotope nuclear reaction network. The results are used to calculate synthetic spectra and light curves, which are then compared with observations of SNe Ia. We also perform binary evolution calculations to determine the number of SNe Ia involving ONe WDs relative to the number of other promising progenitor channels.

Results. The ejecta structures of our simulated detonations in sub-M-Ch, ONe WDs are similar to those from CO WDs. There are, however, small systematic deviations in the mass fractions and the ejecta velocities. These lead to spectral features that are systematically less blueshifted. Nevertheless, the synthetic observables of our ONe WD explosions are similar to those obtained from CO models.

Conclusions. Our binary evolution calculations show that a significant fraction (3-10%) of potential progenitor systems should contain an ONe WD. The comparison of our ONe models with our CO models of comparable mass (similar to 1.2 M-circle dot) shows that the less blueshifted spectral features fit the observations better, although they are too bright for normal SNe Ia.

The white dwarf's carbon fraction as a secondary parameter of Type Ia supernovae

Context. Binary stellar evolution calculations predict thatChandrasekhar-mass carbon/oxygen white dwarfs (WDs) show a radiallyvarying profile for the composition with a carbon depleted core. Manyrecent multi-dimensional simulations of Type Ia supernovae (SNe Ia),however, assume the progenitor WD has a homogeneous chemicalcomposition.
Aims: In this work, we explore the impact ofdifferent initial carbon profiles of the progenitor WD on the explosionphase and on synthetic observables in the Chandrasekhar-mass delayeddetonation model. Spectra and light curves are compared to observationsto judge the validity of the model.
Methods: The explosion phaseis simulated using the finite volume supernova code Leafs, which isextended to treat different compositions of the progenitor WD. Thesynthetic observables are computed with the Monte Carlo radiativetransfer code Artis. Results: Differences in binding energies ofcarbon and oxygen lead to a lower nuclear energy release for carbondepleted material; thus, the burning fronts that develop are weaker andthe total nuclear energy release is smaller. For otherwise identicalconditions, carbon depleted models produce less 56Ni.Comparing different models with similar 56Ni yields showslower kinetic energies in the ejecta for carbon depleted models, butonly small differences in velocity distributions and line velocities inspectra. The light curve width-luminosity relation (WLR) obtained formodels with differing carbon depletion is roughly perpendicular to theobserved WLR, hence the carbon mass fraction is probably only asecondary parameter in the family of SNe Ia.
Tables 3 and 4 are available in electronic form at http://www.aanda.org

Laser Driven Neutron Sources: Characteristics, Applications and Prospects

The basics of laser driven neutron sources, properties and possible applications are discussed. We describe the laser driven nuclear processes which trigger neutron generation, namely, nuclear reactions induced by laser driven ion beam (ion n), thermonuclear fusion by implosion and photo-induced nuclear (gamma n) reactions. Based on their main properties, i.e. point source (< 100 μm) and short durations (< ns), different applications are described, such as radiography, time-resolved spectroscopy and pump-probe experiments. Prospects on the development of laser technology suggest that, as higher intensities and higher repetition rate lasers become available (for example, using DPSSL technology), laser driven methodologies may provide neutron fluxes comparable to that achieved by accelerator driven neutron sources in the near future.

Unraveling the Dust Formation Process in R Dor

Using both dynamical and chemical modelling, we derive an accurate abundance profile for the molecule SiO in the stellar wind of R Dor, an O-rich AGB star. SiO plays a key role in the dust formation process in O-rich AGB stars. This method will be applied to additional molecules, with the aim to achieve a detailed overview of the molecular abundance pattern in the wind of R Dor.