Abstract Multiple parts in gas exchange system of 2-stroke heavy fuel aircraft engine with poppet valves lead to complicated manufacturing and inefficient assembly. Besides, real experimental optimization wastes lots of time and cost due to the increased valve parameters. To address the above issues, the paper proposes a digital twin (DT)-driven optimization method with several DT modules for the system to virtually simulate and optimize the parameters, performance and manufacturing with data interaction and recorded. The DT modules receive real-time feedback data from manufacturing measurements and performance tests to conduct the correction throughout the optimization process. The results demonstrate that the virtual engine model with feedback and correction is quite precise and credible compared with test results, and iterative calculation for optimal parameters is performed efficiently. With the guidance of virtual manufacturing, real manufacturing and assembly are arranged more reasonably and efficiency has been promoted. Real-world test found both power and gas exchange performance improved about 4% under various engine speeds and loads, which verified the effectiveness of DT-driven optimization. The study achieves the integration between virtual and real worlds of the system performance and manufacturing, which facilitates the development of aircraft engine smart manufacturing.

中文翻译：

---

二冲程重油飞机发动机换气系统数字孪生驱动优化

摘要 带提升阀的二冲程重油飞机发动机换气系统零件多，制造复杂，装配效率低。此外，由于阀门参数的增加，真正的实验优化浪费了大量的时间和成本。针对上述问题，本文提出了一种数字孪生（DT）驱动的优化方法，该方法具有多个DT模块，可通过数据交互和记录对系统的参数、性能和制造进行虚拟仿真和优化。DT 模块接收来自制造测量和性能测试的实时反馈数据，以在整个优化过程中进行修正。结果表明，与测试结果相比，带有反馈和校正的虚拟发动机模型相当精确和可信，并且有效地执行最优参数的迭代计算。在虚拟制造的引导下，真实的制造和装配更加合理，效率得到了提升。实际测试发现，在各种发动机转速和负载下，动力和气体交换性能均提高了约4%，验证了DT驱动优化的有效性。该研究实现了系统性能与制造的虚实结合，促进了飞机发动机智能制造的发展。实际测试发现，在各种发动机转速和负载下，动力和气体交换性能均提高了约4%，验证了DT驱动优化的有效性。该研究实现了系统性能与制造的虚实结合，促进了飞机发动机智能制造的发展。实际测试发现，在各种发动机转速和负载下，动力和气体交换性能均提高了约4%，验证了DT驱动优化的有效性。该研究实现了系统性能与制造的虚实结合，促进了飞机发动机智能制造的发展。