Font Size: a A A

Research On Aerodynamic Robustness Optimization Of Centrifugal Compressor Based On Kriging Model

Posted on:2021-04-21Degree:MasterType:Thesis
Country:ChinaCandidate:Z WangFull Text:PDF
GTID:2392330614453713Subject:Engineering
Abstract/Summary:
Centrifugal compressors are of compact structure,light weight and high single-stage supercharging ratio,which are widely used in aerospace,transportation,energy and other fields.Due to its complicated working environment,the aerodynamic boundary conditions such as the speed,temperature and back pressure of centrifugal compressors usually fluctuate randomly.At the same time,At the same time,due to factors such as manufacturing process dispersion and blade pollution,impellers inevitably have geometric errors.The coupling effect of these uncertain factors causes the actual performance of the compressor to deviate from the design value and even the performance of the system is extremely unstable.And the high-speed air flow and strong turbulence inside the impeller also cause serious compressor noise problems.In view of the above problems of aerodynamic performance fluctuations and noise of centrifugal compressors,this paper carries out aerodynamic robustness and noise performance optimization research of centrifugal compressors under the conditions of random aerodynamic boundaries and impeller geometric errors,providing theoretical basis and design for centrifugal compressors and other impeller mechanical designs and applications reference.The main research contents of this article are as follows:1.Based on the 3D flow field numerical analysis,Latin hypercube test design and Kriging surrogate model,the aerodynamic performance response model of the centrifugal compressor is constructed;the influence of the impeller geometry parameters on the aerodynamic performance of the compressor is analyzed;the degree of correlation between aerodynamic performance and structural parameters is quantified.2.Using the non-intrusive probability configuration point method to analyze the impact of impeller speed fluctuations on the aerodynamic performance of the compressor and clarify the sensitivity of aerodynamic performance to speed fluctuations;Based on numerical analysis,non-intrusive probabilistic configuration point method and Kriging surrogate model,a random aerodynamic response model of a centrifugal compressor is established;combined with a multi-objective optimization algorithm,the robustness optimization of the compressor performance under the condition of aerodynamic boundary fluctuations is performed.3.Considering the coupling effect of the impeller geometric error and random speed,a multi-source uncertain aerodynamic response model of the impeller is established based on the Kriging surrogate model and Monte Carlo method,and the degree of fluctuation of the aerodynamic performance of the centrifugal compressor is analyzed.To reduce the size of the optimization target,taking the pressure ratio mean and standard deviation weighting functionof pressure ratio and isentropic efficiency as the optimization objective,the aerodynamic performance and robustness optimization model of the centrifugal compressor impeller under multi-source uncertain conditions are established,and the multi-objective genetic algorithm is used.4.The Proundman-Lilley broadband sound source model is used to analyze the noise of the centrifugal compressor,and the sound power level distribution law of the flow field in the centrifugal compressor is obtained.Based on the Kriging surrogate model,the aerodynamic performance and noise response model of the centrifugal compressor is established,and the influence of the changes in the geometry parameters of the impeller on the pressure ratio and sound power level is analyzed.Considering the aerodynamic and acoustic characteristics,the multi-objective optimization of the aerodynamic robustness and noise performance of the impeller is carried out under the uncertain conditions.
Keywords/Search Tags:centrifugal compressor, aerodynamic performance, noise, uncertainty, robustness optimization
Related items