Optimization of the working process of the axial uncooled turbine using 3D Navier-Stokes solver

The article presents the results of optimization of the working process of the four-stage uncooled low pressure turbine developed and tested in NASA. The aim of the optimization was to increase efficiency of the turbine on the main operation mode by changing the form of rotor and nozzle blades. For the nozzle vanes in the optimization process changed the angles of the sections, the shape of the midline sections, to make sections in the axial and circumferential directions. For the rotor blades of changing the angles of the cross sections. Numerical model of the turbine was set up in the software NUMECA AutoGrid 5 with the estimated net orderly. Three numerical models with varying thickness have been created. The numerical model was verified by comparing the calculated turbine characteristics with the experimental data. As a result of the verification, it was shown that the established numerical model of the working process of the turbine allows to predict tendencies flow characteristics, although it has a quantitative discrepancy with the experimental data. Also, the numerical model provides grid convergence was chosen. To optimize the workflow used the methodology based on the combined use of a program complex mathematical multi-criteria optimization IOSO and software complex computational fluid dynamics NUMECA Fine / Turbo. Parametric change of rotor blades and nozzle blades was carried out in a specially designed Profiler program. During system optimizations have been set restrictions on the position of the working point of the turbine: the working fluid flow rate should differ by no more than ± 0,5% from the reference value; the degree of decompression should differ by no more than ± 0,5% from baseline. Two optimization problem was solved with the use of computational mesh with varying density. One problem with the lowest density, the other with the mesh, ensure the implementation of the convergence of grid conditions. Both problems have shown to achieve the same result - increasing the efficiency by 0.8%. comparative analysis of the basic version and an optimized turbine was performed.