基于支持向量机的非线性系统故障诊断

提出了联想度的概念，并由此设计出一种自组织模糊CMAC（SOFCMAC）及其学习算法，证明了SOFCMAC能以任意精度对非线性特性一致逼近。该网络具有学习速度快，逼近精度高等特点，用该SOFCMAC作为非线性系统观测器而生成残差，通过支持向量诊断器得到故障检测与诊断结果。对某型歼击机的结构故障进行诊断，仿真结果表明了该方法的有效性。

Support Vector Machine Based Fault Diagnosis for Nonlinear Dynamics Systems

A concept of association degree is proposed and a self organizing fuzzy CMAC and its learning algorithm are presented based on CMAC. The nonlinear approximations provided by the SOFCMAC can be made arbitrarily accurate. The proposed network is characterized by fast learning, accurate approximation etc. SOFCMAC is then used as an observer for nonlinear systems to generate residual. The diagnostic results can be obtained by feeding the residual into the support vector machine based diagnostic tool. The proposed method is applied to the structure fault diagnosis for certain fighter aircraft. The simulation results show the effectiveness of the proposed method.