一种基于物理信息核函数神经网络的扩散型动态数据溯源方法
By introducing a physical information kernel function as an activation function into the neural network, a physical information kernel function neural network is constructed. This solves the problems of low efficiency and poor accuracy of traditional methods in long-term heat and mass transfer data tracing, and realizes fast and accurate dynamic data tracing. It is suitable for fire origin analysis, pollutant tracing, and drug molecule delivery path inversion.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HOHAI UNIV
- Filing Date
- 2024-05-08
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional physical information neural networks are inefficient and inaccurate when tracing the source of heat and mass transfer over long periods, making it difficult to simulate reverse heat and mass transfer problems.
A physical information kernel function is introduced as the activation function of a neural network. A physical information kernel function neural network is constructed. By combining the boundary and final conditions of the diffusion equation, a loss function is constructed by modifying the source term. The neural network is then trained to trace the heat and mass transfer dynamic data at any historical moment.
It enables rapid and accurate tracing of heat and mass transfer dynamic data at any historical moment, overcoming the computational bottleneck of traditional methods, improving accuracy and efficiency, and is suitable for fire origin analysis, pollutant tracing, and drug molecule delivery path inversion.
Smart Images
Figure CN118335246B_ABST