Numerical simulation of friction stir-assisted incremental forming with synchronous bonding of heterogeneous sheet metals

Proceedings of the 27th Conference on Computer Aided Architectural Design Research in Asia (CAADRIA) [Volume 2]

“…Hence, it is also necessary to consider alternative strategies for detecting room regions. (Zeng et al, 2019); (c) Semantic segmentation (Kalervo et al, 2019); (d) Instance segmentation (Wu et al, 2020); (e) Graph neural network-based segmentation (Song and Yu, 2021) The proposed framework (Figure 2) can be decomposed into 8 steps: 1) the floor plan image is first pre-processed to improve performance of the segmentation models; 2) the mask of the floor plan is extracted using region adjacency graph extraction algorithm (Song and Yu, 2021) for later usage; 3) semantic segmentation is conducted to obtain initial segmentation of walls, doors and windows; 4) instance segmentation is conducted to obtain wall, door, window and stair instances; 5) the segmentation outputs from step 3 and 4 are used to smooth the wall boundaries and close up wall gaps; 6) pixel-level arithmetic operation is conducted based on the floor plan mask and other architectural elements to extract the room regions; all segmentation results are then fused and vectorized; 7) the region adjacency graph is extracted from the vectorized segmentations; 8) the region adjacency graph is distilled to acquire simpleand multi-attributed adjacency graphs. Using the proposed framework, a customized large-scale attributed adjacency graph dataset is constructed using floor plan images retrieved in bulk.…”

“…These methods all require large data sets. Numerical simulation of ISF (Cai et al ,2020) can simulate the machining process and save the time and cost of data collection. However, the simulation calculation of ISF is difficult, and realizing feedback in real time still remains a challenge.…”

“…Figure 1. Testing results: (a) Original floor plan image; (b) Semantic segmentation(Zeng et al, 2019); (c) Semantic segmentation(Kalervo et al, 2019); (d) Instance segmentation(Wu et al, 2020); (e) Graph neural network-based segmentation(Song and Yu, 2021) …”

Interfacial quality prediction model for Al/steel sheets during friction stir–assisted double-sided incremental forming with synchronous bonding

“…Where Tmelting is the melting temperature of AA5052, and fitting parameters 𝛽, κ vary between 0.04 ~ 0.06, 0.65 ~ 0.75 [27], respectively. To properly describe the heat response in FS-DSIF&SB process, our previous simulation work [28] has revealed that increased ratio of step down and forming angle (i.e. S/𝛼) has a positive effect on the peak temperature of loading area in the set parameters range.…”

Interfacial Quality Prediction Model for Al/Steel Sheets During Friction Stir Assisted Double-Sided Incremental Forming with Synchronous Bonding

“…Where Tmelting is the melting temperature of AA5052, and fitting parameters , κ vary between 0.04 ~ 0.06, 0.65 ~ 0.75 [27], respectively. To properly describe the heat response in FS-DSIF&SB process, our previous simulation work [28] has revealed that increased ratio of step down and forming angle (i.e. S/ ) has a positive effect on the peak temperature of loading area in the set parameters range.…”