一种数据与知识联合驱动的可解释海温预测方法及装置
By employing a data and knowledge-driven approach, multi-scale and frequency domain features of ocean surface temperature are extracted. By combining the ocean surface heat balance equation and physical loss function, the limitations of traditional methods in terms of computational power and the lack of interpretability in machine learning are addressed, thus achieving highly accurate and interpretable ocean surface temperature prediction.
CN118536077BActive Publication Date: 2026-07-17TIANJIN UNIV
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TIANJIN UNIV
- Filing Date
- 2024-05-14
- Publication Date
- 2026-07-17
Smart Images
Figure CN118536077B_ABST
Abstract
本发明公开了一种数据与知识联合驱动的可解释海温预测方法及装置，方法包括：采用ASPP提取海表温度的空间全局与局部特征，采用傅里叶变换提取频域特征，并与空间全局与局部特征拼接，形成海表温度的时空融合特征；根据海洋表面热平衡方程进行辐射通量数据的耦合建模，获取各通量数据间的相互依赖关系，通过3D‑CNN实现特征融合，形成辐射通量的多变量耦合特征；使用PINN作为预测模型，将海表温度的时空融合特征、辐射通量的多变量耦合特征以及海洋动力学参数进行通道拼接形成多模态特征，将多模态特征输入至时空Transformer，通过残差相加构建海洋表面温度预测的输出值。装置包括：处理器和存储器。本发明及时发现海洋表面温度异常变化，避免人员财产的损失。
Need to check novelty before this filing date? Find Prior Art