时间矢量推进方法及其在叶轮机非定常流分析中的应用

为提高周期性非定常流动的求解效率,将非定常计算的初值问题转换为边值问题,提出了时间矢量推进方法,并将该方法应用于叶轮机多排的非定常流动分析中。通过对两排对转风扇进行非定常仿真,并对比双时间步计算方法的计算结果,讨论了新方法的计算效率,研究了该方法对排间干扰捕捉的准确性和可靠性。得到了以下结论:在物理时间步长相等的情况下,新方法与双时间步方法的计算结果基本一致,且计算时间约为双时间步方法的1/8;时间矢量推进方法能够很好捕捉排间的势扰动、熵扰动和涡扰动以及主流和扰动之间的非线性作用;时刻样本数较少会使时间矢量推进方法捕捉到的非定常变化幅值变小,且无法解析时间尺度较小的非定常流动现象。

Time Vector Marching Method and Its Application on Analyzing Unsteady Flows within Turbomachinery

To improve the solution speed of periodic unsteady flows, the Time Vector Marching(TVM) method has been proposed by converting the initial values problem of unsteady solution to a boundary value problem and then applied on unsteady flows within turbomachinery. The comparisons between the TVM and Dual Time Steppeng(DTS) methods were carried out on a counter-rotating fan. The efficiency, accuracy and reliability of the new method on analyzing rows interaction were discussed. Results show that with the same physical time step, the TVM solution is nearly the same with that of the DTS solution, while the time cost of the new method is only one eighth that of the DTS method; The TVM method could capture the potential, entropy and vorticity perturbations and the nonlinear effects between the mean flow and unsteady flow well; For the TVM, less time samples used would lead to small amplitude of unsteady vibrations and the unsteady flows in small time scale would not be resolved.