A turbine blade dynamic stress field prediction method based on hybrid data training
The method for predicting the dynamic stress field of turbine blades by training with mixed data combines numerical calculations and experimental measurement data to construct a deep learning model, which solves the problems of accuracy and coverage in the prediction of dynamic stress field in existing technologies and achieves fast and high-precision prediction of dynamic stress field.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XI AN JIAOTONG UNIV
- Filing Date
- 2026-04-28
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies are insufficient to accurately predict the full-field distribution of dynamic stress on turbine blades. Strain gauge measurement methods have low survival rates at measurement points under harsh conditions, affecting blade safety and failing to cover complex operating conditions. Furthermore, the accuracy of numerical calculations is difficult to match that of experimental measurements.
A hybrid data training method is adopted, which combines numerical model calculation data of blade dynamic stress and experimental measurement data. A dynamic stress field prediction model is constructed through deep learning. The model is trained and fine-tuned using a joint loss function. By integrating high-value sparse experimental data with low-cost and extensive simulation data, a fast and high-precision prediction of the dynamic stress field is achieved.
It enables rapid and accurate prediction of the dynamic stress field of turbine blades, with strong adaptability, high prediction accuracy, fast response speed, high feasibility of on-site deployment, and low long-term operation and maintenance costs.
Smart Images
Figure CN122413318A_ABST