Feedback control of laser forming using flattening simulations for error determination

Research output: Contribution to journalConference article in JournalResearchpeer-review

Abstract

Laser forming is a non-contact forming method. By control of different process parameters, different forming mechanisms can be achieved. The forming mechanisms allows for bending, compression and buckling. Process control to ensure correct combination and usage of each forming mechanism is still an open issue. Present work is an attempt to implement a feedback control loop using measurements of the current shape between laser scan pass. This work presents an initial case study used to evaluate a feedback control loop. A flattening model is used to determine the difference in strains between a target shape and the current shape. The current shape is updated between laser scan paths. The case study examines different shapes including the dome, saddle and v-bend shapes. By examining different shapes that require different forming mechanisms, this work takes steps toward formulating process design principles for use in feedback control of laser forming. The results show decent agreement with the target shape and the formed shape for the v-bend. As the cases increase in complexity, requiring a combination of forming mechanisms, the disagreement increases. The analysis of the process indicates that the boundary conditions of the flattening model and the design of the controller could be improved to reduce tolerances.

Original languageEnglish
Article number012093
JournalIOP Conference Series: Materials Science and Engineering
Volume651
ISSN1757-8981
DOIs
Publication statusPublished - 2019
Event38th International Deep Drawing Research Group Annual Conference - Enschede, Netherlands
Duration: 3 Jun 20197 Jun 2019

Conference

Conference38th International Deep Drawing Research Group Annual Conference
Country/TerritoryNetherlands
CityEnschede
Period03/06/201907/06/2019

Fingerprint

Dive into the research topics of 'Feedback control of laser forming using flattening simulations for error determination'. Together they form a unique fingerprint.

Cite this