首页> 外文会议>International Conference on Materials Processing and Characterisation >Prediction of Forming Limit Diagram for Ti-6A1-4V Alloy Using Artificial Neural Network
【24h】

Prediction of Forming Limit Diagram for Ti-6A1-4V Alloy Using Artificial Neural Network

机译:用人工神经网络预测Ti-6a1-4V合金的形成限位图

获取原文

摘要

In sheet metal industries, the ability to predict and avoid failures, such as necking, fracture and wrinkling are of great importance. It is important to work within the safe strain region to avoid these failures. The forming limit diagram (FLD) is the most appropriate tool to obtain the safe strain region for every sheet metal in different strain conditions and ration. In this paper, significance of important process parameters namely, punch speed, blank holder pressure (BHP) and temperature on FLD of a Ti-6A1-4V alloy are investigated. Taguchi technique was employed to identify the influence of these parameters on major strain and minor strain. The finite element model of deep drawing process has been built up and analyzed using Dynaform version 5.6.1 with LS-Dyna version 971 as solver. The simulations have been performed in Taguchi order to obtain the values of major strain and minor strain. Furthermore, the values of major and minor strain is predicted using artificial neural network (ANN) considering input parameters as punch speed, blank holder pressure (BHP) and temperature and normalized distance of a cup. Forming limit curve is drawn by using Keeler's equation considering the properties of Ti-6A1-4V at various forming temperatures. It is observed that the predicted strain values are in good agreement with the experimental data.
机译:在钣金行业中,预测和避免故障的能力,如颈缩,裂缝和皱纹是非常重要的。重要的是在安全应变区域内工作,以避免这些故障。成形限位图(FLD)是最合适的工具,用于在不同应变条件和配油中获得每个金属金属的安全应变区域。本文研究了重要过程参数的重要性,即Ti-6a1-4V合金的FLD冲压速度,坯料压力(BHP)和温度的FLD。用Taguchi技术鉴定这些参数对主要应变和次要菌株的影响。使用DYNAFORM 5.6.1版使用DYNAFORM 5.6.1版本为单位绘图过程971作为求解器,建立了有限元模型。已经在Taguchi进行了模拟,以获得主要应变和次要应变的值。此外,使用人工神经网络(ANN)预测主要和次要应变的值,考虑输入参数作为冲压速度,坯料支架压力(BHP)和温度和杯子的归一化距离。考虑到各种成形温度下的Ti-6a1-4v的性质,通过使用柯尔的方程来绘制形成极限曲线。观察到预测的应变值与实验数据很好。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号