基于试飞数据的航空发动机部件特性修正

Component Characteristic Correction of Aero-engine based on Flight Test Data

研究了一种发动机部件特性修正方法,提出通过对相似发动机结构的通用稳态模型与试飞数据相匹配来获取专用发动机稳态模型。通过对发动机模型的分析,使用优化算法对各部件特性的流量、压比、效率等参数进行调整,经过修正后的仿真模型在设计点及非设计点的输出与试飞数据的相对误差小于2%,计算精度可满足工程需要。在参数优化方法上比较了粒子群算法(PSO)与进化粒子群算法(EPSO),结果表明,EPSO算法在收敛速度和精度上比PSO算法更为优秀,在处理多变量复杂问题时有较好的寻优能力。

In this paper, a method for the component characteristic correction of aero-engine is studied. The steady-state model of a dedicated engine is obtained by matching the common steady-state model of a similar engine with the test data. Through the analysis of the engine model, the optimization algorithm is used to adjust the flow, pressure ratio, efficiency and other parameters of each component. The relative error between the output of the modified simulation model and the test data is less than 2% at the design point and off design point, and the calculation precision meets the engineering requirement. The particle swarm optimization （PSO） algorithm is compared with evolutionary particle swarm optimization （EPSO） algorithm on the basis of parameter optimization. The results show that EPSO algorithm is superior to PSO algorithm in convergence speed and accuracy, and it can be used to calculate multi-variable complex problems.