Session: 05-01-04 Applied CFD

Paper Number: 68137

Start Time: August 11th, 12:50 PM

68137 - Optimization of a Turbine Rotor Profile by Means of the Use of Cfd and Generic Algorithms 

The paper describes a new optimization strategy for computationally expensive design optimization problems of turbomachinery, combined with design of experiments (D-optimal design), response surface models (RSM), genetic algorithm (GA), and Computational Fluid Dynamics. Latin Hypercube Sampling (LHS) and Latin Hypercube Design (LHD) was carried out for the airfoil design. The quadratic response surface model was used to approximate the relationships between the design variables and the approximation function. The genetic algorithm was applied to the response surface model to perform a global optimization to obtain the optimum design. The above method was applied to the optimization design of HL profile used in the BR715-C130 Engine. The object was to maximized the adiabatic efficiency, and reduce the diffusivity factors using a weight function. As a result of this optimization, the adiabatic efficiency was successfully increased by 1.002%. It was found that the strategy of this paper provides a reliable design optimization method for turbomachinery blades at reasonable cost. This work was done in the Laboratorio de Ingeniería Térmica e Hidráulica Aplicada (LABINTHAP, Laboratory of Thermal and Hydraulic Applied Engineering) from the Instituto Politécnico Nacional, Escuela Superior de  Ingeniería Mecánica y Eléctrica (Faculty of Mechanical and Electrical Engineering), campus Zatenco in Mexico City.

Presenting Author: Miguel Toledo Velázquez Instituto Politécnico Nacional

Authors:

Miguel Toledo Velázquez Instituto Politécnico Nacional
Francisco Eduardo Chavolla Alcalá General Electric Querétaro
Mónica Toledo García Instituto Politécnico Nacional