基于改进PCA的导弹装备健康表征参数提取方法

针对导弹装备健康状态信息复杂且相互交融、健康表征参数难以提取的问题, 提出一种基于改进主成分分析(principal component analysis, PCA)的装备健康状态低维敏感表征参数的确定方法。该方法先开展装备扩展故障模式及影响分析, 构建初始高维特征参数集, 再利用改进PCA对参数集进行降维处理, 在最大化高维表征参数全局特征方差的目标下, 提取出非线性表征参数子集。将该方法应用到导弹舵机健康评估实验中, 使用故障注入模拟设备进行验证。结果表明, 采用所提方法提取的健康表征参数对舵机健康状态识别准确率高, 说明所提方法在提取导弹装备健康表征参数中具有明显的优越性。

Extraction method of missile equipment health characterization parameters based on improved PCA

Aiming at the problems of complex and integrated missile equipment health status information and difficult to extract health characterization parameters, a method for determining low-dimensional sensitive characterization parameters of equipment health status based on improved principal component analysis (PCA) is proposed. Firstly, the equipment extensional failure mode and effect analysis are carried out to construct the initial high-dimensional characteristic parameter set. Then, the improved PCA is used to reduce the dimension of the parameter set. Under the goal of maximizing the global characteristic variance of the high-dimensional characterization parameters, the nonlinear characterization parameters subset is extracted. The proposed method is applied to the health evaluation experiment of missile steering gear, and verified by the fault injection simulation equipment. The results show that the health characterization parameters extracted by the proposed method have high recognition accuracy for the health status of steering gear, indicating that the proposed method has obvious advantages in the extraction of missile equipment health characterization parameters.