多源信息融合的航空部附件状态退化预测

Predicting the Status Degradation of Aviation Accessories Based on Multi- source Information Fusion

针对航空部附件状态退化预测问题,提出了一种经验模态分解(EMD)和自组织特征映射(SOFMs)量化误差模型相结合的部附件退化趋势预测方法。采集部附件多个特征指标的状态监测信息,运用EMD提取包含微弱信号的特征信息,并消除部分噪声干扰;运用SOFMs实现多源传感器信息融合,并建立最小量化误差(MQE)模型,量化部附件运行状态,以实现部附件的状态退化预测。通过对某型航空陀螺仪的仿真验证表明,EMD-SOFMs量化误差模型能够有效、准确地提取陀螺仪状态信息,融合量化陀螺仪运行状态,实现陀螺仪的状态退化预测。

For predicting the status degradation of aviation accessories, the method combining empirical mode decomposition （ EMD） and selforganizingfeature maps （ SOFMs） quantization error model is put forward. The status monitoring information about multiple feature indexes ofaccessories is collected; by using EMD,the feature information containing weak signals is extracted,and partial noise interference is eliminated.Multi - source sensor information fusion is realized through SOFM, and the minimum quantization error （ MQE） model is established; to quantifythe running status of accessories and implement prediction of status degradation of accessories. A certain type of aviation gyroscope is selectedfor conducting simulation and verification. Examples of analysis and verification show that EMD - SOFMs quantization error model doeseffectively and precisely extract the status information of gyroscope, fuse and quantify operating status of gyroscope, and implement the predictionof status degradation for gyroscope.