학술논문

Springback Prediction and Forming Accuracy Control of Micro W-bending Using Support Vector Machine
Document Type
Conference
Source
2019 6th International Conference on Frontiers of Industrial Engineering (ICFIE) Frontiers of Industrial Engineering (ICFIE), 2019 6th International Conference on. :23-27 Sep, 2019
Subject
Engineering Profession
Support vector machines
Training
Predictive models
Grain size
Kernel
Annealing
Artificial neural networks
springback
micro forming
prediction
support vector machine
forming accuracy
Language
Abstract
Accurately predicting and effectively controlling of the springback of micro W-bending is evidently important to improve its dimensional accuracy and forming quality. For the first time, micro W-bending is proposed in this paper. Firstly, the experimental design method based on the I-optimal criterion was adopted to test the four factors (foil thickness, gain size, punch frequency and punch displacement) affecting the forming accuracy, and 56 sets of data were obtained. Secondly, support vector machine (SVM) was used to predict the springback of micro W-bending. Afterwards, the prediction values obtained by support vector machine were compared with the experimental results. It is showed that the predicted values fit well with the experimental values. The minimum relative error is 0.3%. Finally, some measures are addressed, providing references to control and improve the forming accuracy of the W-shaped micro-bent parts.