Abstract:Blade is one of the core components of turbine machinery. The reliability of blade is directly related to the normal operation of plant unit. However, with the increase of blade length and flow rate, non-linear effects such as finite deformation must be considered in strength computation to guarantee enough accuracy. Parallel computation is adopted to improve the efficiency of classical nonlinear finite element method and shorten the blade design period. So it is of extraordinary importance for engineering practice. In this paper, the dynamic partial differential equations and the finite element method forms for turbine blades under centrifugal load and flow load are given firstly. Then, according to the characteristics of turbine blade model, the classical method is optimized based on central processing unit + graphics processing unit heterogeneous parallel computation. Finally, the numerical experiment validations are performed. The computation speed of the algorithm proposed in this paper is compared with the speed of ANSYS. For the rectangle plate model with mesh number of 10 k to 4000 k, a maximum speed-up of 4.31 can be obtained. For the real blade-rim model with mesh number of 500 k, the speed-up of 4.54 times can be obtained.

Study on Finite Deformation Finite Element Analysis Algorithm of Turbine Blade Based on CPU+GPU Heterogeneous Parallel Computation