Abstract
Swept blades are widely utilized in transonic compressors/fans and provide high load,high through-flow,high efficiency,and adequate stall margin.However,there is limited quantitative research on the mechanism of the effect of swept blades on the flow field,resulting in a lack of direct quantitative guidance for the design and analysis of swept blades in fans/compressors.To better understand this mechanism,this study employs a reduced-dimensional force equilibrium method to analyze more than 1500 swept cascades data.Results verify that circumferential fluctuation terms are responsible for inducing radial migration in the inlet airflow field of the swept blade,resulting in variations in the incidence angle and consequently leading to changes in the characteristics of the swept blade.Thus,a combination of simple functions and machine learning is utilized to model the circumferential fluctuation terms and quantify the sweep mechanism.The prediction accuracy of the model is high,with coefficient of determination greater than 0.95 on the test set.When the model is applied in a meridional flow analysis program,the calculation accuracy of the program for the incidence angle is improved by 0.4° and 0.6° at the design and off-design conditions respec-tively,compensating for the program's original deficiencies.Meanwhile,the model can also provide quantitative guidance for the design of swept blades,thereby reducing the number of design itera-tions and improving design efficiency.
基金项目
National Natural Science Foundation of China(52376021)
NETL
NSTL
万方数据