摘要
为解决长期以来在产品造型设计中缺乏系统化、自动化的设计方法,以汽车侧面造型为例,利用贝塞尔曲线描述汽车造型,确立汽车造型变量,从而可使用数值来表现造型。利用自组织映射神经网络建立模拟消费者的认知机制,分析各造型类别的数值差异,并计算其共同的特性,提取造型特征。该文的研究结果可建立一套能自动提供消费者对产品认知模式的方法,藉以加速设计的流程,辅助设计师有效地进行产品造型设计。
Due to the difficulty in introducing the systematic and automatic methods to product shape design, a side shape design of an automobile is described as an example to show the possible solution by using new design method presented in this paper. The method is firstly to employ bezier curve methodology to quantify shape of an automobile and to establish the shape variables; secondly to simulate the consumer recognizing mechanism by means of an unsupervised learning neural network ＂Self-Organization Mapping＂; thirdly to separate the features of the design shape into different categories and analyze their variable differences accordingly, and finally to derive their common features. The results can be used to build a decision supporting system to shorten the design period and assist the designer to perform the product design effectively.
出处
《工程图学学报》
CSCD
北大核心
2007年第3期113-117,共5页
Journal of Engineering Graphics
