Multi-stages, multi-responses optimisation of wire arc additive manufacturing parameters using Taguchi method. (2022)

• Record Type: Journal Article
• Authors: Wani, Z.K.; Abdullah, A.B.; Jaafar, N.A.; Hussain, Z
• Is Part Of: Materials today. Volume 66:Part 5(2022)
• Page Start: 2660
• Page End: 2664
• Publication Date: 2022
• Subjects: WAAM -- Mild steel -- Multi-stages optimization; AM Additive Manufacturing -- WAAM Wire Arc Additive Manufacturing -- ANOVA Analysis of Variance -- EDM Electrical Discharge Machining -- GMAW Gas Metal Arc Welding -- OA Orthogonal Array -- CMM Coordinate Measuring Machine; Materials science -- Congresses -- Periodicals; 620.1
• DOI: 10.1016/j.matpr.2022.06.490
• Languages: English
• ISSNs: 2214-7853

Abstract

Abstract: Wire arc additive manufacturing (WAAM) is a technique that combines automated welding with direct deposition of a melted material in a 3D profile. The technique allows for quick and flexible deposition of a material in various applications. This paper presents a multi-stage, multi-response optimisation of WAAM parameters. The performance of the deposited metal in WAAM is influenced by several factors, including welding parameters (e.g., speed and feed rate) and process parameters (e.g., deposition sequence) and defined based on the application and desired profile quality. In this study, optimisation is divided into two stages on the basis of the input parameters. Stage 1 covers voltage, wire speed and travel speed, and Stage 2 includes number of layers and delay time. The Taguchi approach is used to determine the effects of input parameters on the response of the volume and height per layer of the 3D profile for Stages 1 and 2, respectively. The height per layer of the 3D profile is evaluated based on the total height of the specimens divided by the number of layers printed. Experiments are conducted to achieve an L9 orthogonal array by using two different factorial approaches, namely, 3**3 factorials for Stage 1 and 3**2 factorials for Stage 2. ANOVA results show that the combination of A2B3C1 has the highest volume of the 3D profile, and A1B3 presents the highest height per layer value. The conformation tests for both stages are omitted here because Taguchi's optimal output matched the optimum value achieved in the parametric test. Furthermore, the full factorial experiment was completed for the Stage 2 experiment.