Research on Aerodynamic Optimization Method of Multistage Axial Compressor under Multiple Working Conditions Based on Phased Parameterization Strategy

Abstract

Multistage axial compressor is the key component of aeroengine and gas turbine to realize energy conversion. In order to avoid the “curse of dimensionality” problem in the global optimization process of AL-31F four-stage low-pressure compressor under multiple working conditions, an optimization method based on phased parameterization strategy is proposed. The method uses the idea of “exploration before exploitation” for reference and divides the optimization process into two phases. In the first phase, the traditional parametric modification method based on stacking line is adopted; in the second phase, the full-blade surface parametric modification method with significant low-dimensional characteristics is adopted. Based on the improved artificial bee colony algorithm, a multitask concurrent optimization system is built on the supercomputing platform, and the engineering optimization solution is obtained within 91 hours. The optimization results are as follows: under the condition of meeting the constraints, the adiabatic efficiency is increased by 0.3% and the surge margin is 4.0% at the design speed; the adiabatic efficiency is increased by 0.8% and the surge margin is 2.3% at the off-design speed. These results verify the usefulness and reliability of the optimization method in the field of aerodynamic optimization of a multistage axial flow compressor.