基于支持向量机的航空发动机辨识模型

针对航空发动机具有强非线性、时变性的特点以及采用常规神经网络辨识时存在的局部较小,过学习等问题,提出了一种基于支持向量机的通用发动机模型辨识方法。该方法基于现代统计学习理论,采用结构风险最小化准则,保证了网络具有很强的推广特性,通过求解凸二次规划确保网络结构全局最优化自动生成。本文采用实测到的发动机飞行记录数据作为训练样本,利用回归型支持向量机建立了发动机的辨识模型,研究结果表明该方法的辨识精度较高,鲁棒性、容错性较好,具有较大的实用价值。 

A Novel Aero-Engine Identification Model Based on Support Vector Machines

Because of the strong non-linearity and time-varying properties,and the existence of local minima,overlearning of conventional neural networks, a novel aero-engine identification model is introduced.The model is based on modern statistical learning theory,and Structure Risk Minimization (SRM) principle.It has very good generalization ability.By solving a quadratic convex programming problem, a global optimum can be automatically found.The identification model was established based on support vector regression machines using the real flight recorded data as learning samples.The results show that this method reveals high precision,good robustness and fault-tolerant ability. It provides a general way for aero-engine model identification.