Efficient design optimization method and system for super-large variable turbomachinery and application

Abstract

The invention discloses an efficient design optimization method and system for an ultra-large variable impeller machine and application. The method comprises the following steps: setting an ultra-large variable impeller machine parameterization method; setting a CFD calculation model according to the design working condition of the optimization design; uniform initial point adding is carried out in a global range, and an evaluation value is calculated; a high-dimensional PCE fitting model is established globally; decomposing a high-dimensional problem into a plurality of low-dimensional sub-problems; performing independent proxy model optimization on all the sub-problems, and adding a global PCE model as a low-precision data source into an HK model in the optimization process to realize knowledge migration; combining all sub-problem optimization results to obtain a group of global samples and evaluating the global samples; updating the global model and repeating the steps until an optimization stopping condition is met; according to the design method, the variable number of the design can be increased on the premise that the sample number can be accepted, so that the degree of freedom of the design is expanded, the parallel expansion capability is very high, the design time is short, and the calculation efficiency is high.

Description

technical field [0001] The invention belongs to the field of impeller machinery optimization, and in particular relates to a method, system and application of an efficient design optimization method and system of an ultra-large variable impeller machinery. Background technique [0002] In recent years, in the field of turbomachinery, automatic optimal design methods have been more and more widely used. The automatic optimal design method needs to manually determine a design space that can be parameterized, and each sample in the space corresponds to a specific geometric design scheme. Taking the results of computer simulation calculations as the optimization target, with the help of specific optimization algorithms, turbomachinery components with excellent performance can be automatically designed. The use of automated design methods can effectively reduce the need for designers' experience, and complete the design quickly and with high quality. [0003] Optimization algor...

AI Technical Summary

Problems solved by technology

But in actual engineering optimization, the number of samples is limited by computing power and time

(2) The premise of searching and adding new samples is the accuracy of the proxy model. In the case of low accuracy of the high-dimensional proxy model, it cannot guide the subsequent addition of points

(3) The time complexity of agent model establishment and optimization search increases sharply with the increase of sample size and dimension

(4) The Gaussian kernel function used in the Kriging model is implemented based on Euler distance, but there is a situation where Euler distance fails in high-dimensional space

Images