Method for extracting time sequence features of spacecraft system health state and fault prediction
By combining modal decomposition and graph attention mechanisms with TCN and Transformer methods, the problem of mining deep spatiotemporal coupling features in spacecraft telemetry data was solved, enabling accurate prediction and efficient early warning of early faults, improving prediction accuracy and reducing latency.
CN122412905APending Publication Date: 2026-07-17XIAN TRANSPORT CONTROL INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by
Patent Information
- Application Number
- CN202610472805.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-10
- Publication Date
- 2026-07-17
Smart Images
Figure CN122412905A_ABST
Abstract
本发明公开了一种航天器系统健康状态的时序特征提取与故障预测方法，属于航天器故障预测技术领域。该方法包括：获取原始多源遥测序列，利用模态分解算法将其分解为本征模态函数，并按中心频率重构为趋势分量与波动分量；将各传感器定义为图节点，基于关联关系动态构建特征图拓扑结构，利用图注意力机制提取空间维度上的耦合特征；将空间耦合特征并行输入TCN支路与Transformer支路。本发明通过模态分解剥离工况噪声、动态图注意力捕获空间耦合、TCN与Transformer并行融合多尺度时序特征，有效解决了航天器遥测数据中早期微弱故障难以提取的问题，显著提高了故障预测精度并缩短了预警时延，适用于卫星、深空探测器等航天器的在轨健康管理。
Need to check novelty before this filing date? Find Prior Art