Search

A parallel full implicit Navier-Stokes solver based on weakly imposed wall boundary condition is proposed. Fir s t, a weak boundary procedure for viscous solid wall is implemented on the unstructured node-centered finite volume method. Upon this procedure, the corresponding viscous Jacobian is derived. Compared to the traditional strongly imposed wall boundary condition, the application of weakly imposed boundary condition eases the derivation. To address the issues of convergence and positivity preserving difficulties with regard to the Spalart-Allmaras (S-A) model, an unconditionally positive-convergent (U PC) implicit scheme for the S-A model using weakly imposed wall boundary condition is presented, which is based on the construction of a special implicit matrix to ensure both the convergence of the S-A model and the positivity of the turbulence working variables. The linear equations resulted from the full implicit time integration and the UPC scheme are solved using multicolor Gauss-Seidel iteration (MCGS) and MPI is employed for parallelization. The proposed approach is tested by simulating several two-dimensional and three-dimensional cases. Results from the numerical simulations demonstrate that; 1) with larger penalty strength parameter, the results of weak boundary conditions tend to those of strong boundary conditions; 2) the weak boundary conditions satisfy flux consistency for force computation automatically, and therefore no special treatment is required. [ABSTRACT FROM AUTHOR]

Linearized Euler equations with the source terms are solved by finite difference method based on the SNGR method. The spatial discretization is conducted by the dispersion-relation-preserving (DRP) scheme. The time integration is implemented by low-dissipation and low-dispersion Rung-Kutta schemes. The non-reflecting far-field boundary conditions are utilized. The noise of the two-dimensional cavity with length to depth ratio 9 : 1 is predicted at different Mach numbers under the subsonic, transonic and supersonic conditions, and the results agree with the experimental data. [ABSTRACT FROM AUTHOR]

Discontinuous Galerkin (DG) finite element method was adopted for the numerical approximation of the Reynolds-averaged Navier-Stokes(RANS) equations with the Spalart-Allmaras turbulence model. In order to make the solver robust, the original turbulence model equation were modified accordingly. Furthermore, high order approximation of the real solid boundary was used and several layers of curved meshes were constructed to avoid inconsistent mesh cross-overs. For the computation of the distance of each quadrature point to the nearest curved wall boundaries, a fast straightforward numerical method was proposed. The DG discretization of the RANS equations were demonstrated for turbulent flows past a NACA0012 airfoil and RAE airfoil based on hybrid mesh. Numerical results indicate that highly accurate solutions can be obtained with the modified turbulent equation on coarse curved hybrid mesh. [ABSTRACT FROM AUTHOR]

Exposure to ambient particulate matter (PM) links with a variety of respiratory diseases. However, compared with coarse particles (PM 10 ) and fine particles (PM 2.5 ), submicrometer particles (PM 1.0 ) may be a more important indicator of human health risks. In this study, the cytotoxic and genotoxic effects of PM 1.0 samples from Shanghai were examined using A549 cells, and compared with the effects of PM 2.5 , to better understand the health effects of PM 1.0 in this area. The PM 1.0 and PM 2.5 samples were characterized for morphology, water-soluble inorganic ions, organic and elemental carbon, and metal elements. The cytotoxicity of PMs was measured using cell viability and cell membrane damage assays. The genotoxic effects of PMs were determined using the comet assay, and DNA damage was quantified using olive tail moment (OTM) values. The physicochemical characterization indicated that PM 1.0 was enriched in carbonaceous elements and hazardous metals (Al, Zn, Pb, Mn, Cu, and V), whereas PM 2.5 was more abundant in large, irregular mineral particles. The biological results revealed that both PM 1.0 and PM 2.5 could induce significant cytotoxicity and genotoxicity in A549 cells, and that exposure to PM 1.0 caused more extensive toxic effects than exposure to PM 2.5 . The greater cytotoxic effects of PM 1.0 can be attributed to the combined effects of size and chemical composition, whereas the genotoxic effects of PM 1.0 may be mainly associated with chemical species.