BP神经网络在盒形件坯料外形预测中的应用

【摘要】 确定坯料外形是拉深件工艺设计中的一项重要工作。本文应用基于BP算法的神经网络能较好处理非线性问题的特点，建立起能描述工艺参数（材料参数、几何参数等）和毛坯外形的映射关系的神经网络模型；结合研究对象的特点，提出了确定最优网络结构的方法。计算结果与有限元预测结果进行了比较。更多还原

【Abstract】 Determination of blank shape is an important task in deep drawing process. On account of that BP neural network solves nonlinear problems well,a BP neural network is set up to map the relation between process parameters (material parameters and tools’ geometry parameters) and blank shape. A method is proposed to design optimum network architecture corresponding to the problem in this paper. The results are compared with those from FEM.更多还原