Abstract: In order to solve the problem of insufficient accuracy of the traditional piece wise linear on-board aero-engine model，nonlinear remainders of Taylor expansion for each model outputs are taken into account. And a new on-board steady-state aero-engine model considering these nonlinear items is proposed and established. For the purpose of estimating real engine component deviations，a component deviation estimation model based on kalman filter is designed. Thus a new on-board adaptive aero-engine model is completed and it takes the proposed steady-state aero-engine model as the core and combines the component deviation estimation model. On the basis of the adaptive engine model，necessary simulations about one component and multi-components deviation estimation and performance seeking control are carried out，respectively. The simulation results show clearly that the error of the adaptive engine model is within one percent，and the model has smaller amounts of data storage with quicker time responses.

Key words: Steady-state variable model；Adaptive model；Aero-engine；Performance seeking control；Deviation estimation

摘要： 为了解决传统的机载发动机分段稳态线化模型精度不足的问题，在发动机稳态线化模型中引入了模型各输出参数泰勒展开中的非线性余项，建立了考虑非线性余项的机载发动机稳态模型。为了估计真实发动机部件蜕化情况，建立了基于Kalman滤波的发动机部件性能蜕化估计模块。以考虑非线性余项的发动机稳态模型为核心，结合性能蜕化估计模块构建了机载发动机自适应模型。针对所建立的机载发动机自适应模型，进行了单部件及多部件蜕化参数估计以及最小油耗性能寻优控制模式的仿真。仿真结果表明，考虑非线性余项的机载发动机自适应模型误差在1%以内，且具有优化耗时少 ，建立模型样本数据需求量小的特点。

关键词: 稳态模型；自适应模型；航空发动机；性能寻优控制；蜕化量估计