Online Control of Deposited Geometry of Multi-layer Multi-bead Structure for Wire and Arc Additive Manufacturing
A robotic wire and arc additive manufacturing (WAAM) system with active vision sensing capability was implemented to improve the geometry accuracy of multi-layer multi-bead structures. Width and height of the deposited bead were online detected and the ac
- PDF / 2,904,485 Bytes
- 9 Pages / 439.37 x 666.14 pts Page_size
- 117 Downloads / 220 Views
)
Harbin Institute of Technology, West Da-Zhi Street 92, Harbin 150001, People’s Republic of China [email protected]
Abstract. A robotic wire and arc additive manufacturing (WAAM) system with active vision sensing capability was implemented to improve the geometry accu‐ racy of multi-layer multi-bead structures. Width and height of the deposited bead were online detected and the accuracies of the designed sensing approaches are 0.25 mm and 0.1 mm, respectively. A double-input-double-output controller with two subsystems was designed to ensure the uniformity of bead width and bead height. The bead width was adjusted by a single neuron self-learning PI controller, while the bead height was controlled based on a rule-based engine. The experi‐ mental results indicated that (i) accuracy of the deposited bead width is controlled in 0.5 mm, (ii) accumulation effect of the bead height deviations has disappeared when multiple layers are overlapped, and (iii) the surface finish of each layer was controlled to be flat as well. Keywords: Wire and arc additive manufacturing · Multi-layer multi-bead structure · Active vision sensing · Bead geometry control
1
Introduction
Additive manufacturing (AM) is a technology to fabricate parts by depositing material on a layer-by-layer basis. AM is capable to realize complex structure design and fabri‐ cation, and reduces material and energy at the same time [1, 2]. Wire and arc additive manufacturing (WAAM) is an alternative AM approach for metallic part deposition. The heat source of WAAM is welding arc, while the fed material is wire. It has been proved that WAAM requires lower cost on implementations and fabricates parts with higher efficiency in comparison with other metallic AM technics [3]. The research activities of WAAM regarding advanced material shaping, microstructure and mechan‐ ical performance analysis of the deposited part, and path optimization techniques have been reported in the literature succession [4–6]. In general, WAAM contains a planning phase and a deposition phase. In the planning phase, a 3D model of the target part is sliced into multiple layers. For each layer, the deposition path is designed based on the contour of the layer. In addition, manufacturing parameters are deduced according to the planned bead geometries of the layers.
© Springer Nature Singapore Pte Ltd. 2018 S. Chen et al. (eds.), Transactions on Intelligent Welding Manufacturing, Transactions on Intelligent Welding Manufacturing, DOI 10.1007/978-981-10-5355-9_7
86
Q. Han et al.
However, geometries of the deposited bead are susceptible to be influenced by the diverse heat sinking conditions and thermo-mechanical coupling effects, which might cause deviations on the bead geometries to the planned values [7, 8]. When the deviations accumulate, the subsequent deposition process cannot advance strictly as planned. Therefore, the WAAM process has the fatal shortcomings of poor continuity and low accuracy. Improving the accuracy of bead geometry is one of the research challenges in WAAM. To this end,
Data Loading...
