Machine Learning for the Prediction of Edge Cracking in Sheet Metal Forming Processes
- MACHINE LEARNING FOR THE PREDICTION OF EDGE CRACKING IN SHEET METAL FORMING PROCESSES -
The chapter "Machine Learning for the Prediction of Edge Cracking in Sheet Metal Forming Processes" resulted from the SAFEFORMING project, led by Toolpresse between 2017 and 2020.
- The book "Machine Learning and Artificial Intelligence with Industrial Applications" was published this year and consists of a compendium of information and case studies that provide a comprehensive view on computer learning and artificial intelligence and their industrial applications.
- The chapter "Machine Learning for the Prediction of Edge Cracking in Sheet Metal Forming Processes" resulted from the SAFEFORMING project, led by Toolpresse between 2017 and 2020.
- The objective of the project was to evaluate the performance of several computational learning algorithms in the prediction of defects in the metal stamping process, namely the occurrence of cracking.
- Seven different unique classifiers and two types of ensemble models (majority voting and stacking) were used to make predictions, based on a dataset generated from the results of two types of mechanical tests: the uniaxial tensile test and the hole expansion. The performance evaluation was based on four metrics: accuracy, recall, precision and F-score, with the F-score considered the most relevant. The best performances were achieved by the ensemble majority voting models. The ROC curve of a majority model was also evaluated in order to confirm the predictive capabilities of the model. Overall, ML algorithms are able to predict the occurrence of edge cracking satisfactorily.
20/04/2022