基于蒙特卡罗法的航空发动机空气系统稳态算法优化

针对目前航空发动机空气系统稳态算法中收敛性依赖初值的问题，将蒙特卡罗方法与流体网络法综合应用到空气系统可压缩流体一维网络计算中，提出了一种新的计算方法Monte Carlo-Fluid Network（MC-FN）。该方法将空气系统简化为由节点和元件组成的网络，借助蒙特卡罗方法获得空气系统内各节点压力分配，再根据空气系统中各元件流阻特性和换热特性计算流量、温度。计算中通过将游动次数比较少的蒙特卡罗方法的计算结果作为流量残差法节点压力、温度的初始值，实现快速求得精确收敛解。与流量残差算法相比，MC-FN方法计算精度不变，收敛速度提升了66.5%；与线性求解法相比，MC-FN方法的计算精度提升了25.2%，收敛速度提升了43.8%。

Optimization of Steady State Algorithm for Aero Engine Air System Based on Monte Carlo Method

To solve the initial value dependence occurred in the current aero-engine air system steady-state algorithm, a new calculation method Monte Carlo-Fluid Network(MC-FN), which combined Monte Carlo (MC) method and fluid network method, was applied to solve 1D network calculation of compressible fluid in air system. The air system was simplified as a fluid network comprised of nodes and elements. The pressure distribution of each node was calculated by the MC method, and then the mass flow rate and temperature was calculated based on the flow resistance characteristics and heat transfer characteristics of each element in the air system. An accurate convergence solution can be quickly obtained by using the calculation results of the MC method with relatively few walking times as the initial value of the node pressure and temperature of the flow residual method. Compared with the flow residual method, the calculation accuracy of the MC-FN method is not improved, but the convergence speed is increased by 66.5%; Compared with the linear solution method, the calculation accuracy of the MC-FN method is improved by 25.2%, and the convergence speed is increased by 43.8%.