Development of Precise Prediction System of Welding Deformation for Large-scale Structure Using GPGPU Computing

2015 Fiscal Year Final Research Report

• Project/Area Number: 25420873
• Research Category: Grant-in-Aid for Scientific Research (C)
• Allocation Type: Multi-year Fund
• Section: 一般
• Research Field: Naval and maritime engineering
• Research Institution: Osaka Prefecture University
• Principal Investigator: Shibahara Masakazu 大阪府立大学, 工学(系)研究科(研究院), 准教授 (20350754)
• Co-Investigator(Kenkyū-buntansha): ITOH Shinsuke 大阪府立大学, 大学院工学研究科, 客員准教授 (50535052) ; MURAKAWA Hidekazu 大阪大学, 接合科学研究所, 教授 (60166270)
• Project Period (FY): 2013-04-01 – 2016-03-31
• Keywords: GPGPU / 溶接変形 / 大規模解析 / 高速解析 / 熱弾塑性解析 / 船体構造 / 理想化陽解法FEM / ルートギャップ

Outline of Final Research Achievements

In this research, a new analysis system based on Idealized Explicit FEM using GPU was developed. Additionally, massively parallel processing is introduced to this system for high-speed computation. The proposed system is method is effective in large-scale analysis. Using this system, it's possible to predict the influence of several factors related to welding procedure such as tack welds, behavior of root gap, misalignment in an assembly process.
By developing this system, the prediction of deformation during welding assembling process in ship building becomes to be possible. Since this system can be applied to not only welding deformation problem in ship building but also welding residual stress problem of various structure. This means that the proposed system is promising in the future.

Free Research Field

船舶海洋工学