论文部分内容阅读
周向收敛型动压式指尖密封是一种可用于航空发动机的非接触式柔性气封。根据周向收敛型动压式指尖密封的工作性能要求,提出密封性能优化数学模型。采用人工神经网络和遗传算法相结合的优化方法,通过建立反向传播(BP)神经网络确定设计变量与目标函数之间的隐含关系,再利用遗传算法对周向收敛型动压式指尖密封进行结构参数优化,获得了性能较优的周向收敛型动压式指尖密封结构形式。通过对具有优化结构的密封在转子启动工作阶段和转子受到周期性径向激励情况下的流固耦合动态性能仿真,验证了周向收敛型动压式指尖密封优化结果的有效性。本文工作可为设计性能良好的周向收敛型动压式指尖密封提供一定的理论依据。
The circumferentially convergent dynamic pressure fingertip seal is a non-contact, flexible gas seal for aero-engines. According to the working performance requirement of circumferential convergence dynamic pressure fingertip seal, a mathematical model of sealing performance optimization is proposed. The artificial neural network and genetic algorithm are combined to optimize the method. The BP neural network is used to determine the implicit relationship between the design variables and the objective function. The genetic algorithm is used to analyze the circumferential convergence dynamic pressure fingertip Sealed to optimize the structural parameters, get better performance of the circumferential convergence of dynamic pressure fingertip seal structure. Through the simulation of the fluid-solid coupling dynamic performance of the seal with optimized structure during the start-up phase of the rotor and the cyclic radial excitation of the rotor, the effectiveness of the optimization results of the circumferential convergence dynamic pressure fingertip seal is verified. The work in this paper can provide a theoretical basis for the circumferential design of the dynamic convergence of the dynamic pressure fingertip seal.