A method and device for springback compensation of a stamped part based on a machine learning model
By using a machine learning model-based method, finite element simulation and convolutional neural network training data sets to establish a rebound compensation big data model, the problems of high mold development cost and long cycle in traditional methods are solved, and fast and effective mold surface compensation and part dimensional accuracy are achieved.
Patent Information
- Application Number
- CN202410211644.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-02-27
- Publication Date
- 2025-10-17
- Estimated Expiration
- 2044-02-27
AI Technical Summary
Traditional methods of compensating for springback of stamped parts are highly dependent on technicians' process experience and stamping simulation technology, resulting in high mold development costs and long cycles, making it difficult to effectively ensure the dimensional accuracy of stamped parts.
A method based on machine learning model is adopted to establish a rebound compensation big data model through finite element simulation prediction and data acquisition. The sample data set is trained using the convolutional neural network algorithm to directly predict the optimal rebound compensation result, reducing the trial and error iterative process.
It achieves fast, effective and low-cost mold surface compensation, improves the dimensional accuracy of stamping parts, reduces dependence on process experience, and shortens the mold production cycle.