Probabilistic Design Optimization of Frequency Dispersion for Rotating Blades

In this paper, a probabilistic design optimization method based on finite element method is proposed to calculate the variability of design parameters subject to a specified dispersion of natural frequencies of rotating blades. The element stiffness and mass matrices are derived using a two-stage finite element method and numerical integration. Based on the perturbation technology, the sensitivity of the frequencies, as well as relationship between the frequency dispersion and the coefficient of variability (CV) of the design parameters can be obtained. Such sensitivity information is then used to convert the probabilistic design optimization problem into a deterministic optimization problem. Two case studies are given to illustrate the proposed method. From the results, it is concluded that rotation of blade changes the sensitivity of CV to the design parameters considered, and using the proposed method can transform the probabilistic constraints to deterministic constraints.Copyright © 2010 by ASME