Machine Learning for Predicting Fracture Strain in Sheet Metal Forming

Machine learning models are built to predict the strain values for which edge cracking occurs in hole expansion tests. The samples from this test play role of sheet metal components be manufactured, often associated with a uniaxial tension stress state at critical edges components. For construction models, dataset was obtained experimentally rolled ferritic carbon steel sheets different qualities and thicknesses. Two types tests were performed: tensile In test, yield stress, strength, maximum load elongation after fracture determined rolling transverse directions. occurs, determined. It is intended that can based on knowledge data. machine algorithms used Multilayer Perceptron, Gaussian Processes, Support Vector Regression Random Forest. traditional polynomial regression fits 2nd order function also comparison. shown learning-based predictive outperform method; particular, Processes found best enable most robust models.