Research and Reviews - International Journals

Research & Reviews: Journal of Engineering and Technology

Menu

ISSN: 2319-9873

Reach Us +44 7456 035580
All submissions of the EM system will be redirected to Online Manuscript Submission System. Authors are requested to submit articles directly to Online Manuscript Submission System of respective journal.

Artificial Neural Networks applied to Bending Processes, Identification of Material Parameters and Processing Variables


Webinar on 2nd International Congress on AI and Machine Learning

February 15, 2022 | Webinar

Daniel Jacome da Cruz

Institute of Science and Innovation in Mechanical and Industrial Engineering, Portugal

ScientificTracks Abstracts: RRJET

Abstract

Machine learning approaches have reawakened interest as a result of increased data availability and processing capabilities, which is becoming a significant trend in a variety of domains and applications. As a result, sheet metal forming and other manufacturing processes are using such oportunity, having in mind the efficiency and control of the many parameters involving the processing and material characterization. One of the objectives of the present work is to study the applicability of machine learning algorithms on bending procedures, exploring the modeling capabilities of Artificial Neural Networks (ANNs) to solve two distinct problems directly related to bending processes. The first problem incorporates a new methodology to characterize the hardening behavior of a material based in a efficient and simple procedure of a three-point bending test. The developed ANN considers as input the information obtained in a three point bending test and is based in a multi-layer feedforward conventional structure to provide the characteristic parameters of a Swift hardening law. The second problem focuses on the springback problem in sheet metal press-brake air bending, with the objective of predicting the punch displacement required to achieve a desired bending angle, as well as providing additional springback angle information. In both approaches it is proposed to combine the use of a learning tool with a simulation and data generation tool (FEA) in order to train the developed ANN which in turn is validated by experimental results.

Biography

Daniel Jácome da Cruz completed the Master in Mechanical Engineering in setember 2019 granted by Faculty of Engineering of University of Porto. He works in the area of advanced manufacturing processes with emphasis on sheet metal forming. He has published 3 articles in journals and 2 sections of books