ABSTRACT There is significant importance in accurate prediction of axial flow compressor characteristics at different conditions. Different techniques and models have been proposed for performance prediction of axial compressors. Because of its lower computational cost and high speed the meanline algorithm has been widely applied in preliminary design and analysis of axial compressors; however, there are some inaccuracies at design and off‐design conditions since the method relies on empirical correlations, which may be weak when applied to unconventional airfoil types. Applying some modifications to the meanline algorithm could improve its performance for a wider operating range with higher accuracy. This study aims to propose a modification for accuracy enhancement of meanline techniques to obtain characteristics of axial flow compressors at design and different off‐design conditions. For this purpose, three scenarios are considered to modify the models. In the 1st scenario, coefficients are used for the deviation models while the pressure loss was not changed, in the 2nd scenario coefficients are applied for the pressure loss models and the deviation model is used as the base model and in the 3rd scenario, coefficients are used for both models. The coefficients are optimized by use of multi‐objective genetic algorithm. It was found that use of the 3rd scenario leads to the best accuracy. In this scenario, the average absolute error in the estimated isentropic efficiency is reduced from 2.41% to 0.75% at 80% design rotational speed while the improvement in estimated mass flow rate at this speed is relatively minor. However, at design rotational speed after optimization, the average absolute error in the estimated isentropic efficiency drops from 6.35% to 0.96% while the estimated mass flow rate decreases from 4.97% to 0.094%.
Hassanlue et al. (Wed,) studied this question.