Numerical simulation and experimental verification on distortions induced by wire-arc additive manufacturing components and costing analysis / Keval Priapratama Prajadhiana

This thesis focuses on the substrate and part distortion induced by wire arc additive manufacturing (WAAM) which is predicted by means of numerical computation followed by experimental verification. The material used for this study is the commercial S235 low carbon steel and filler material stainless steel SS3161. On analysing the distortion effect by means of numerical computation method, a commercial specialized simulation software Simufact.Welding was used in the development of the numerical model. The development of numerical simulation model started by the geometrical modelling followed by material modelling based on real scanned material using JMATPRO. A mesh sensitivity analysis was executed in order to determine the most optimized element numbers which are followed by the selection of optimized WAAM parameters based on the result of preliminary experiment process. To enlighten the application of method and procedure of numerical computation, this research is divided into two case studies. The first case study is modelling, simulating and fabricating the twenty layers deposition of hollow rectangular WAAM and the second case study is to examine pipe flange WAAM where the substrate and part distortions was be examined. In order to minimize the computational time of numerical WAAM simulation, two method has been proposed. The first method is utilizing the lumping layering of WAAM deposition to minimize the numbers of elements and the second method is by using Inherent Strain Method (ISM) with the help of commercial finite element software MSC Marc/Mentat. A series of comprehensive WAAM experiment was conducted in order to verify the result prediction of numerical simulation which was completed using robotic welding machine ABB IRB 2400/16 and power source KEMMPI Pro Evolution ProMIG 540 MXE where the WAAM trajectory depositions were developed by using RAPID programming in ABB RobotStudio software. The completed WAAM parts undergone substrate removal process and was measured using a 3D scanner GOM Atos Q. A result of investigation shows that the substrate distortion proposed by four different numerical models shows an acceptable range of error when compared to numerical simulation with the average of 10.77% until 20.32%. Once the substrate has been removed, the component distortion comparison is examined by means of modified substrate removal algorithm where the error percentage of the models ranges from 2.96% until 8.05%. The result with optimized parameters and boundary conditions were transferred towards second case study which is a Pipe Flange WAAM component based on ASME B16.5 standard where the development of numerical simulation and fabrication of WAAM parts followed similar procedures with the first case study. Three numerical models which consists of TMM and two different lumping layers