Method for training fatigue life prediction model of grouting connection structure, fatigue life prediction method, device and product
By combining CNN and LSTM, spatial features of grouting connection structures are extracted from CT scan images and fatigue life prediction models are trained using physical constraints. This solves the problem of fatigue life prediction under data constraints, achieves accurate prediction of grouting connection structures, and ensures the safe operation of offshore wind turbines.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JINAN UNIVERSITY
- Filing Date
- 2026-06-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing methods for predicting the fatigue life of grouting connection structures for offshore wind turbines are limited by data acquisition constraints, making it difficult to accurately predict their fatigue life and threatening the operational safety of offshore wind turbines.
A method combining convolutional neural networks (CNN) and long short-term memory networks (LSTM) is adopted to extract spatial feature information from CT scan images of grouting connection structures. The fatigue life prediction model is trained by combining the loss function of physical constraints, so as to achieve accurate prediction of the damage state and fatigue life of grouting connection structures.
With limited data, it is possible to accurately predict the fatigue life of grouting connection structures, reduce reliance on large-scale fatigue testing, and ensure the safe operation of offshore wind turbines throughout their entire life cycle.
Smart Images
Figure CN122416166A_ABST