Scientia Iranica
Volume:17 Issue: 6, 2010

Stay and guide vanes (distributor) are essential parts of a turbine. They are used to control the ow rate and to appropriately transfer the ow momentum to the runner. In this work, ow through the distributor is analyzed. For various Boundary Conditions (BC) and di erent con gurations, threedimensional ows in the distributor of a Francis turbine are evaluated and compared with each other. The numerical simulations were carried out using Fluent software and the results were validated with a GAMM Francis turbine, where the geometry and detailed best eciency measurements were publically available. In these simulations, the ow was assumed to be steady and the e ect of turbulence was included using the k 􀀀 «turbulence model. The study showed that an accurate prediction of velocity and pressure elds through the distributor may be obtained by considering a representative runner chamber with a single passage, including one blade of a stay and guide vane con guration. Furthermore, the corresponding needed computational resources for such an analysis are quite modest.

In this paper, a two-dimensional numerical approach is used to study the e ect of micro combustor height, mass ow rate and external convection heat transfer coecient on the temperature and species mass fraction pro les. A premixed mixture of H2-Air with a multi-step chemistry is used. The transient gas phase energy and species conservation equations result in an Advection-Di usion-Reaction system that leads to two sti systems of PDEs. In the present work, the computational domain is solved through the Strang splitting method, which is suitable for a nonlinear sti system of PDEs. A revised boundary condition for the velocity equation is applied and its e ect on the ow characteristics is investigated. The results show that both convection heat transfer coecient and micro combustor height have a signi cant e ect on the combustion and heat transfer rates in the micro scales. Also increasing the convective heat transfer coecient and decreasing the height and inlet mixture velocity, decreases temperature and active radicals along the micro combustor. In addition, the slip ow and thermal creep boundary conditions in the studied scales have no signi cant e ect on the di erent parameters, but changes slightly the cross section pro les of the temperature. The 2-D numerical results show that the micro combustion must be treated as two dimensional.

In this work, basic design features of Rayan are documented. One of the new design features presented in this work is the way Rayan handles polyhedral grids. Grid de nition is combined with the de nition of the structure of the sparse coecient matrix, thereby releasing a considerable part of the memory used by the grid to store otherwise required faces belonging to the cell part of the connectivity description. The key idea is to use a uniform way for creating the structure of the coecient matrix from the grid connectivity description and to access that data when computing the elements of the coecient matrix. This saving requires many modi cations to the computational algorithm details, which are addressed. Computational method features include a SIMPLE-based pressure-velocity coupling and co-located variable arrangement in which all ow variables are stored at cell centers, and mass uxes are stored on face centers. Also handling convective and di usive uxes is described. The throughput is benchmark validated and shows second order truncation properties, both in time and space.

In this paper, the State Dependent Riccati Equation (SDRE) method is implemented on robotic systems such as a mobile two-links planar robot and a xed 6R manipulator with complicated dynamic equations. Dynamic modelings of both cases are presented using the Lagrange method. Afterwards, the Dynamic Load Carrying Capacity (DLCC), which is an important characteristic of robots, is calculated for these two systems. DLCC is calculated for the prede ned end-e ector path, where motor torque limits and tracking error constraints are imposed for this calculation. For a mobile two-links planar robot, the stability constraint is discussed by applying a zero moment point approach. A nonlinear feedback control law is designed for the fully nonlinear dynamics of two cases using a nonlinear closed-loop optimal control method. For solving the SDRE equation that appears in the optimal control solution, a power series approximation method is applied. DLCC is obtained, subject to accuracy and torque constraints, by applying this feedback control law for the square and linear path of the end-e ector for mobile twolink and a 6R manipulator, respectively. Finally, simulations are done for both cases and the DLCC of manipulators is determined. Also, actual end-e ector positions, required control e orts and the angular position and velocity of joints are presented for full load conditions, and results are discussed

Extensive tests were carried out on a section of a wind turbine blade. The e ect of reduced frequency on the boundary layer transition point of the model oscillating in plunge has been investigated. The spatial-temporal progressions of the transition point and the state of the unsteady boundary layer were measured using multiple hot- lm sensors. The measurements showed that reduced frequency highly a ects variations of the transition point and results in a hysteresis loop in the dynamic transition locations. The dominated frequencies of the boundary layer are found to be a function of the reduced frequency and mean angle of attack.

The air ow field around a surface mounted hemisphere for different velocities is investigated experimentally and numerically. Hot-film anemometry is used in a wind tunnel to analyze the ow structure for different regions of ow domain around the hemisphere. The ow Reynolds number is varied based on the hemisphere diameter and free stream velocities. The air velocity is taken as 5, 7 and 8.5 m/s, with corresponding Reynolds numbers of 35000, 50000 and 64000. To illustrate the ow pattern around the hemisphere, variations of ow velocity at various vertical sections along the midplane, and cross-stream velocities for various surfaces and heights, are measured and plotted. Corresponding numerical solutions are also carried out to visualize the ow reversal, horseshoe vortices and recirculation zones around and downstream of the hemisphere. It is concluded that the ow is highly Reynolds number dependent, especially at the reversed ow region upstream and downstream of the hemisphere. Moreover, the e ect of hemisphere location in the tunnel is studied, and velocity distributions are compared and presented in the wake of the hemisphere for constant free stream velocity.

Small and medium break LOCA accidents at low pressure and under low velocity conditions have been studied in the TTL-2 Thermo-hydraulic Test Loop, experimentally. TTL-2 is a thermal hydraulic test facility which is designed and constructed in NSTRI to study thermal hydraulic parameters under normal operational and accident conditions of nuclear research reactors. A nodalization has been developed for the TTL-2 and experimental results have been compared with RELAP5/MOD3.2 results. The considered accidents are a 25% and 50% cold leg break without emergency core cooling systems. Results show good agreement between experiments and RELAP5/MOD3.2 results. This research provides experimental data for evaluation of thermo hydraulic codes for nuclear research reactors, and veri es that RELAP5/MOD3.2 has a good capability to estimate the thermal hydraulic behavior of low pressure and low velocity thermal hydraulic systems, such as research reactors under steady state and transient conditions.