Inner wrinkling phenomenon is more likely to develop during hydrodynamic deep drawing (HDD) of complicated component-forms due to the higher demand for controlling deformation sequences. Aiming at the problems in co...Inner wrinkling phenomenon is more likely to develop during hydrodynamic deep drawing (HDD) of complicated component-forms due to the higher demand for controlling deformation sequences. Aiming at the problems in control of inner wrinkling for an irregular surface part featured with both concavity and convex, this research proposes an optimal design method of drawbead parameters to change the material flow. According to theoretical analysis of the mechanism of inner wrinkling, optimizing cavity pressure only is unreasonable to form a wrinkle-free deep-drawn part, so semi-circular drawbeads are employed. The effects of layout and height of drawbeads on forming results are discussed, and a process window is established based on evaluation indicators including the anti-wrinkle coefficient and the minimum wall thickness. Experiments are carried out to validate the process window, and the wall thickness and the wrinkle height are measured and compared with numerical findings. The results show that the anti-wrinkle ability of drawbeads weakens with increasing oblique angle and distance from the die center, while the wall thickness increases with increasing oblique angle and distance, and the inner wrinkling can be completely suppressed by drawbeads arranged in zones I and II with optimum penetration.展开更多
针对传统工业生产中二维不规则钣金零件的利用率不高、计算时间较长的问题,提出了以能量函数为载体的BL-NFP(Bottom Left-No Fit Polygon)神经网络算法。该算法是将BL(Bottom-Left)定位算法和临界多边形(No-Fit Polygon,NFP)几何特性相...针对传统工业生产中二维不规则钣金零件的利用率不高、计算时间较长的问题,提出了以能量函数为载体的BL-NFP(Bottom Left-No Fit Polygon)神经网络算法。该算法是将BL(Bottom-Left)定位算法和临界多边形(No-Fit Polygon,NFP)几何特性相结合,同时模拟了钣金零件的排样过程。并采取对待排入零件优先进行面积大小核算、再排入待排物体的方式,并利用Matlab对算法的输出数据和前人所列出的数据结果进行了测试对比。结果表明:BL定位算法能够合理地计算出零件的排入位置,NFP能够有效地解决不规则零件排样利用率小的问题,神经网络算法则能够有效地提高求解速度。针对二维不规则钣金零件的排样问题,与传统神经网络算法相比较,采用该算法缩短了钣金零件下料机器计算最优解40%的时间,并提高了约8%的钣金材料利用率。展开更多
基金supported by Chengdu Aircraft Industrial Corporation
文摘Inner wrinkling phenomenon is more likely to develop during hydrodynamic deep drawing (HDD) of complicated component-forms due to the higher demand for controlling deformation sequences. Aiming at the problems in control of inner wrinkling for an irregular surface part featured with both concavity and convex, this research proposes an optimal design method of drawbead parameters to change the material flow. According to theoretical analysis of the mechanism of inner wrinkling, optimizing cavity pressure only is unreasonable to form a wrinkle-free deep-drawn part, so semi-circular drawbeads are employed. The effects of layout and height of drawbeads on forming results are discussed, and a process window is established based on evaluation indicators including the anti-wrinkle coefficient and the minimum wall thickness. Experiments are carried out to validate the process window, and the wall thickness and the wrinkle height are measured and compared with numerical findings. The results show that the anti-wrinkle ability of drawbeads weakens with increasing oblique angle and distance from the die center, while the wall thickness increases with increasing oblique angle and distance, and the inner wrinkling can be completely suppressed by drawbeads arranged in zones I and II with optimum penetration.
文摘针对传统工业生产中二维不规则钣金零件的利用率不高、计算时间较长的问题,提出了以能量函数为载体的BL-NFP(Bottom Left-No Fit Polygon)神经网络算法。该算法是将BL(Bottom-Left)定位算法和临界多边形(No-Fit Polygon,NFP)几何特性相结合,同时模拟了钣金零件的排样过程。并采取对待排入零件优先进行面积大小核算、再排入待排物体的方式,并利用Matlab对算法的输出数据和前人所列出的数据结果进行了测试对比。结果表明:BL定位算法能够合理地计算出零件的排入位置,NFP能够有效地解决不规则零件排样利用率小的问题,神经网络算法则能够有效地提高求解速度。针对二维不规则钣金零件的排样问题,与传统神经网络算法相比较,采用该算法缩短了钣金零件下料机器计算最优解40%的时间,并提高了约8%的钣金材料利用率。