Improving the accuracy of one-dimensional analysis in axial compressor performance prediction

In the development of compressor, one-dimensional (1D) analysis is indispensable as it can quickly determine the basic parameters and predict characteristics of compressors. To enhance the accuracy of 1D analysis, genetic algorithms and comparison methods are used to obtain the calculation coefficients for deviation angle, work, and efficiency in the empirical models. An eight-stage axial compressor was chosen for optimization and validation. At 90% and 100% design speeds, the 1D results with applying the coefficients for deviation angle, work, and efficiency closely match the characters of the experimental data. The error in choking flow rate and peak efficiency are within 2% when compared to both experimental data and computational fluid dynamics (CFD) results. From 50% to 80% design speeds, the maximum error of 1D results with applying the coefficients for deviation angle and work in choking flow rate and peak efficiency are within 1.6% when compared to both the experimental data and CFD results. Additionally, applying these coefficients in 1D analysis also improves the prediction accuracy for compressors with variable inlet guide vanes(VIGV).