Doctor of Philosophy
School of Mechanical, Materials, Mechatronic and Biomedical Engineering
In the last ten years, research interests in various aspects of the Wire Arc Additive Manufacturing (WAAM) processes have grown exponentially. More recently, efforts to integrate an automatic quality assurance system for the WAAM process are increasing. No reliable online monitoring system for the WAAM process is a key gap to be filled for the commercial application of the technology, as it will enable the components produced by the process to be qualified for the relevant standards and hence be fit for use in critical applications in the aerospace or naval sectors. However, most of the existing monitoring methods only detect or solve issues from a specific sensor, no monitoring system integrated with different sensors or data sources is developed in WAAM in the last three years. In addition, complex principles and calculations of conventional algorithms make it hard to be applied in the manufacturing of WAAM as the character of a long manufacturing cycle. Intelligent algorithms provide in-built advantages in processing and analysing data, especially for large datasets generated during the long manufacturing cycles. In this research, in order to establish an intelligent WAAM defect detection system, two intelligent WAAM defect detection modules are developed successfully. The first module takes welding arc current / voltage signals during the deposition process as inputs and uses algorithms such as support vector machine (SVM) and incremental SVM to identify disturbances and continuously learn new defects. The incremental learning module achieved more than a 90% f1-score on new defects. The second module takes CCD images as inputs and uses object detection algorithms to predict the unfused defect during the WAAM manufacturing process with above 72% mAP. This research paves the path for developing an intelligent WAAM online monitoring system in the future. Together with process modelling, simulation and feedback control, it reveals the future opportunity for a digital twin system.
Li, Yuxing, Machine Learning Based Defect Detection in Robotic Wire Arc Additive Manufacturing, Doctor of Philosophy thesis, School of Mechanical, Materials, Mechatronic and Biomedical Engineering, University of Wollongong, 2021. https://ro.uow.edu.au/theses1/1408
FoR codes (2008)
0913 MECHANICAL ENGINEERING
Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.