[en] This work is devoted to the constitutive modelling of hydromechanical behaviours of semi-brittle geomaterials. In the first part, the attention is mainly paid on the hydromechanical behaviours of initially isotropic materials, especially on the influence of desiccation process on the mechanical responses. An elastoplastic damage model is proposed in unsaturated condition, a generalized effective stress concept is used for poro-plastic coupling. Damage by microcracks is coupled with plastic deformation. The proposed model is applied to two typical semi-brittle materials: the cement and argillite on different saturation degrees of saturation. The mechanical response of a concrete beam in different saturation conditions is also analysed. In order to avoid the problem of strain localisation, a nonlocal formation is proposed. In the second part, two coupled elastoplastic damage models are proposed for the description of strongly anisotropic sedimentary materials. In the first model, in order to describe the inherent anisotropy of the material, a scalar anisotropy parameter is introduced using the concept of fabric tensor. Using this parameter, an anisotropic model is formulated taking into the account influence of loading orientation. Plastic deformation is then coupled with induced damage by growth of microcracks. In purpose of studying the coupling phenomenon between the inherent and induced anisotropies during the loading process, a model based on the discrete approach is constructed in the following. In the framework of plastic discrete approach, the macroscopic plastic deformation and material degradation are considered as the result of frictional sliding along weakness planes distributed randomly in the material. For each weak plane, damage evolution law and plastic flow are formulated with the framework of irreversible thermodynamics. A series of numerical simulations on the Argillite of Tournemire are carried out to verify the capacities of proposed models. In the last chapter, the proposed discrete model is extended to poro-elastoplastic modelling. Based on a series of numerical simulation, an analysis of hydromechanical behaviour of anisotropic materials is realised. (author)

[en] Using the 5.8 x 10⁶ J/ψ events collected by the MARK 3 experiment on the SPEAR e⁺e^- storage ring at SLAC, a mass independent amplitude analysis of the J/ψ → γ K_sK_s and J/ψ → K⁺K^- decays is presented. For K bar K systems having mass less than 2 GeV/c², the efficiency-corrected spherical harmonic moments of the J/ψ joint decay angular distribution are measured. Fits are then performed in each independent mass interval in order to extract the underlying helicity amplitude structure; amplitudes describing K bar K systems of spin zero and spin two are included simultaneously. For the first time, a large spin zero component in the θ(1720) mass region is observed; consistent results are obtained for the data samples corresponding to the individual decay modes. This structure is attributed to the production of an S wave resonance, the f₀(1710), of mass and width M = 1710 ± 10 MeV/c², Γ = 186 ± 30 MeV/c², respectively, with branching fraction Bf(J/ψ → γf₀, f₀ → K bar K) = (6.47 ± 1.14 ± 0.84) x 10^-4. A small amount (∼24%) of spin two component in this mass region cannot be ruled out with the present statistics. These results revise the previous conclusion that the θ(1720) is a spin two resonance, a result obtained on the basis of spin hypothesis tests, which assumed that either pure spin zero or pure spin two states contribute in this mass region, but not both. The previous measurements of the f₂' (1525) have been refined in the present analysis because of the simultaneous inclusion of spin zero and spin two amplitudes in the fit

[en] A mass independent amplitude of the K bar K and π⁺π_- systems (M < 2 GeV/c²) produced in J/ψ radiative decay is presented. For the first time a large spin zero component in the θ(1720) mass region is observed, with all data samples analyzed. A small amount of spin two component in this mass region for the K bar K data samples is not ruled out with the present statistics. This study reveals, also for the first time, the production of the f_o(1400) in the π⁺π^- channel, and refines previous measurements of the f₂(1270) and f'₂(1525). 13 refs., 2 figs., 2 tabs

[en] The TCP end-to-end congestion control plus RED router queue management can be modeled as a discrete-time dynamical system, which may create complex bifurcating and chaotic behavior. Based on the basic features of the TCP-RED model, we propose a time-dependent delayed feedback control algorithm to control chaos in the system by perturbing the accessible RED parameter p_max. This method is able to stabilized a router queue occupancy at a level without knowing the exact knowledge of the network. Further, we study the situation of the presence of the UDP traffic

[en] Highlights: • A novel multi-tank TES for solar-power air conditioning was proposed. • An optimized control strategy of multi-tank TES was proposed. • The number of tanks had little effect on duration time of IHIWT in multi-tank TES. • Longer duration time of IHIWT was obtained after optimized control strategy was adopted. - Abstract: The volume of hot storage tank in sensible thermal energy storage (STES), integrated with solar-powered air conditioning to mitigate the misalignment between supply and demand, is too big which results in great thermal inertia. In this paper, a novel multi-tank thermal energy storage (TES) system for solar-power air conditioning, with the advantages of quick temperature rising and outstanding ability to fit variable weather, was proposed to mitigate the contradictions between demand and supply. In order to investigate effect of the number of tanks (NT) on system response time (SRT), total energy grade of water in tanks and temperature variations in charging and discharging periods, a series of experiments of TESs with different number of tanks (total volume of tanks is constant) were conducted under the same solar radiation conditions. Results show that the energy grade of water in tanks is improved through dividing thermal energy storage system into several tanks with different priorities of charging and discharging, and SRT decreases by 6.95 h in four-tank TES, compared with one-tank TES. Furthermore, the duration time of ideal heating inlet water temperature (IHIWT) in entire operational period increases slowly with the number of tanks, because NT has opposite effects on duration time of IHIWT in charging and discharging periods. An optimized control strategy that tanks stored heat successively in charging period and released heat simultaneously in discharging period was proposed. It is notable that the duration time of IHIWT in four-tank TES increases 2.59 h, about 26.67%, after the optimized control strategy is adopted. Therefore, the optimized TES makes the single-effect absorption chiller run at a high COP over a long time without strong variations of COP, compared with other TESs.

[en] Highlights: • An improved model to predict soot formation in a diesel engine has been proposed. • The physical process of PAHs deposition on the particle surface was considered. • Model validation carried out with a complete experimental matrix in two engines. • Good agreement between experimental and simulated results was found. • Particle size distribution and mass concentration in diesel engines are determined. - Abstract: Although much research has been done on soot formation in engines, the mechanisms involved in the process are poorly understood and models of the process are overly simplistic. An improved detailed soot model is developed that couples a reduced diesel surrogate fuel chemical reaction mechanism of n-heptane/toluene, implemented into KIVA-3V2 code, for the numerical investigation of soot formation, mass concentration, and size distribution in diesel engines. This detailed soot model incorporates the effects of soot precursors, including isomers of acetylene and polycyclic aromatic hydrocarbons (PAHs), and the physical processes of PAH deposition on the particle surface, soot formation, and particle surface growth. Compared with experiment results in an optical engine and a single-cylinder diesel engine, the improved detailed soot model was effective: the simulated in-cylinder combustion pressure, heat release rate, and ignition timing were in excellent agreement with the experimental results. The simulated two-dimensional, transient distribution of soot concentration was in good agreement with that obtained by using the two-color method, and the simulated changing trend of soot emission was consistent with the experimental results. Therefore, the detailed soot model can be used to accurately simulate and predict soot emission at different conditions in diesel engines. Furthermore, at the initial stage of combustion, large amounts of small-size soot particles were produced by the pyrolysis reactions and polymerization of the hydrocarbon fuel, and the particle size was in the range of 5–40 nm. At the middle stage of combustion, soot particles continued to grow by particle coagulation, surface growth, and the deposition of PAHs, which greatly expanded the particle size range and caused many large-size particles to be produced in the engine cylinder. At the late stage of combustion, the range of particle size distribution stabilized under the influence of further oxidation reactions, and the particle size ranged from 5 to 20 nm. The details of soot-relevant quantities (e.g., particle size, number concentration, and mass) provide valuable insights into soot formation and oxidation processes in diesel engines

[en] Using a dynamic oblique angle deposition in a flipping rotation mode where the flux incident angle continuously changes during the magnetron sputter deposition, we demonstrated that a stable, biaxial (110)[11^¯0] body center cubic (alpha-phase) tungsten nanorod film can be grown on amorphous substrates. In contrast, we showed that only a fiber-textured, metastable A15 (beta-phase) tungsten nanorod film was obtained using the conventional rotation mode where the oblique incident flux angle was fixed with the substrate rotating around the surface normal. Different flipping rotation speeds were performed to study the effect of dynamic shadowing on texture axis angular dispersion. The sample grown at 0.3 rotations per minute was observed to have the minimum out-of-plane and in-plane orientation dispersions characterized by the reflection high energy electron diffraction surface pole figure technique. The biaxial texture selection under the flipping rotation mode is explained qualitatively. - Highlights: • Stable alpha-phase biaxial W nanorods were produced with dynamic flipping rotation. • Metastable beta-phase W nanorods were grown by stationary oblique angle deposition. • Biaxial stable W nanorods could be used as a buffer layer for semiconductors

[en] Preliminary results of a mass independent amplitude analysis of J/ψ radiative decays into K bar K final states are presented. A large component of spin zero is observed at the f₂/θ(1720) mass region; however, a small spin two component at this mass region cannot be excluded with the present statistics. 9 refs., 7 figs

[en] The M184V mutation in HIV reverse transcriptase (RT) is associated with high-level resistance against the nucleoside inhibitor lamivudine as well as diminished viral replication capacity. We have previously demonstrated that HIV variants containing the M184V mutation were relatively unable to successfully undergo compensatory mutagenesis following deletion of an A-rich loop located upstream of the primer binding site (PBS). To understand the mechanisms involved, we synthesized viral RNA templates containing different compensatory mutations that were emergent during the long-term culture of the A-rich loop-deleted viruses. These templates were then used in cell-free reverse transcription initiation assays and in tRNA primer placement assays performed with either recombinant wild-type RT or recombinant RT containing the M184V substitution. The results showed that the RNA template that contained the A-rich loop deletion was impaired in ability to initiate reverse transcription and that the presence of the M184V substitution in RT amplified this effect. Clearance from pausing at position +3 during synthesis of viral DNA was identified as a sensitive step in this reaction that could not be efficiently bypassed with the M184V mutant enzyme. Increased efficiency of initiation was seen with the deleted RNA templates that also contained mutations identified in the revertant viruses, provided that these mutations facilitated formation of a competent binary tRNA/RNA complex. These findings provide biochemical evidence that initiation of tRNA^Lys3-primed DNA synthesis is an important rate-limiting step in reverse transcription that correlates with viral replication fitness

[en] This paper presents a digital method for the poles identification of an analog system consisting of a preamplifier and a shaping filter in the digital spectroscopy. Based on this method, the result of the simulation of the analog and digital system is modeled in the Matlab and Simulink platform. The influence of noise, the 'ghost zero' and some realization problem is also discussed. The algorithm is based on SVD of the Hankel. With the advantage of accuracy, operating in real-time, numerical stability and requiring no initialization, the method is widely used. (authors)