Fatigue Endurance Optimization of a Steam Turbine Blade Curved Fir-Tree Root

Document Type : Original Research Paper

Authors

1 Mechanical Rotary Equipment Research Department, Niroo Research Institute (NRI), Tehran, Iran

2 Steam Engineering Unit, Turbomachinery Department, MAPNA Turbine Engineering and Manufacturing Company (TUGA), Karaj, Alborz, Iran

10.22084/jrstan.2023.27108.1228

Abstract

In this paper, the guide curve of a fir-tree root is parametrically modeled for the optimization of a newly designed free-standing low-pressure last-stage blade airfoil of a steam turbine. These geometrical parameters include the location of the center point of the curve root centerline and the radius of the root curvature, while the axial length and the pitch are kept fixed. The mechanical safety factor is computed at the upper section of a predefined root profile for each considered design parameter in a commercial finite element software. For high cycle fatigue failure analyses, the maximum alternating stresses are calculated from harmonic responses under the dynamic fluid loading obtained from computational fluid dynamics (CFD) simulation at the maximum continuous rating condition. Preliminary modal analysis is performed to estimate the natural frequencies for the principal modes near the low engine order excitation. On this basis, dynamic stresses are calculated at corresponding frequencies prone to low order engine modes of excitement. Moreover, harmonic response is obtained at nozzle passing frequency for each set of design parameters. In addition, the total strain is computed at the upper section of the root from elastic-plastic analysis under centrifugal loading resulting from 110% of nominal operating speed. Therefore, a multi-objective optimization genetic algorithm is used to optimize the root design parameters with objectives of maximizing safety factors obtained from both strain-based low cycle fatigue and stress-based high cycle fatigue analyses.

Keywords

Journal of Stress Analysis

Articles in Press, Accepted Manuscript
Available Online from 27 December 2023

Files

Share

How to cite

Statistics