复合材料覆盖件及其模具曲面逆向工程造型

利用径向基函数（RBF）神经网络优异的非线性逼近能力，将流线型玻璃钢覆盖件样件外形数据、加工余量、变形数据作为神经网络输入，在神经网络的输出上可以分别得到覆盖件曲面和模具曲面离散数据点。再通过神经网络的输出数据对曲面进行造型。通过在复杂曲面模具造型上的实际应用，证明该方法能够实现产品和工艺装备的并行设计，缩短产品研制周期，提高了设计和生产效率。

Reverse Engineering Styling of Composite Panels and Die Surface

With the surface data of the streamline-formed glass fiber reinforced plastics exterior panels,allowance and deformation data as inputs,utilizing the superior nonlinear approximation capacity of the radial basic function neural network,discrete point data on panel surface and die surface can be obtained from output end of the neural network.Then surface styling is carried out on the basis of output data of the neural network.By actual application in a complex die surface styling,it is proven that this method can realize concurrent design of product design and process design,shorten product development cycle and improve design and production efficiency.