基于边界分割与频域桥接的复杂构件非均应力场重构方法
By using boundary segmentation and frequency domain bridging, complex components are divided into multiple sub-regions. An adapted physical information neural network is constructed and a dynamic link is established in the frequency domain. This solves the problems of efficiency and accuracy in stress field reconstruction of complex components, and realizes high-precision reconstruction of stress field and effective fusion of multi-source data.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- ZHEJIANG UNIV
- Filing Date
- 2026-04-24
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to achieve efficient and high-precision reconstruction of non-uniform stress fields under complex geometric models, especially in key load-bearing components of high-end equipment. Finite element analysis preprocessing is cumbersome and computationally time-consuming, and a single physical information neural network cannot simultaneously fit high-frequency and low-frequency stress distributions. Furthermore, simulation data and measured data are difficult to cross-integrate.
Complex components are divided into multiple sub-regions by boundary segmentation. A physical information neural network embedded with the elasticity mechanism equation is constructed. A frequency domain bridging module is used to establish a dynamic link of the stress field in the sub-region in the frequency domain. Multi-source stress field data are fused through a variable fidelity cascaded neural operator network to achieve high-precision reconstruction of the stress field.
It achieves efficient and high-precision reconstruction of non-uniform stress fields in complex components, eliminates the non-physical step and discontinuity problems in stress field splicing, improves the fidelity of stress field reconstruction and engineering practical value, and meets the needs of rapid iterative design of multiple schemes and online state perception.
Smart Images
Figure CN122088313B_ABST