BP神经网络在程控切纸机整体造型感性设计中的运用  

Application of BP Neural Network in Perceptual Form Design of Programmable Paper Cutting Machine

作　　者：定律 朱彦[2] DING Lu;ZHU Yan(University of Shanghai for Science and Technology,Shanghai 200093,China;Shanghai Dianji University,Shanghai 200245,China)

机构地区：[1]上海理工大学,上海200093 [2]上海电机学院,上海200245

出　　处：《包装工程》2020年第16期135-140,共6页Packaging Engineering

基　　金：上海市哲学社会科学规划项目;上海市教委教师专业发展工程国外访学项目(A1-0217-18-003-05)。

摘　　要：目的运用计算机图形设计和逻辑运算方法建立用户需求与造型设计要素之间的数理联系,对程控切纸机产品进行更科学的造型设计与评价,使设计流程更加严谨。方法以语义差异法和因子分析法归纳用户感性意象评价,以形态分析法解构程控切纸机整体造型设计要素,建立感性评价矩阵,用以训练BP神经网络。结果测试后发现网络正确映射了感性意象和设计要素之间的关系,用该网络模型进行模拟预测,得到关于“高档的—低端的”感性评价最大值和最小值,对应的造型设计要素组合能给设计“高档”风格的程控切纸机提供理性支撑。结论将用户模糊不清的情感转化为定量的数据,弥补原先设计流程中单纯凭设计者主观经验去比较和评价设计方案的不足,为大型机电产品整体造型设计提供科学的设计方法。The work aims to establish the mathematical relationship between user’s demands and form design elements by computer graphics design and logical operation method,so as to make more scientific design and evaluation about the programmable paper cutting machine and enhance the design process more rigorous.User’s perceptual image evaluation was summarized by semantic difference method and factor analysis method,integral design elements of programmable paper cutting machine were deconstructed by morphological analysis method,and perceptual evaluation matrix was established to train BP neural network.The network correctly mapped the relationship between the perceptual image and design elements after test.The maximum and minimum values of perceptual evaluation about high-end and low-end were obtained through the simulated prediction by the network model,and the corresponding combination of design elements provided rational support for the design of high-end style programmable paper cutting machine.The user’s ambiguous emotion is transmuted into quantitative data,which makes up for the deficiency of comparing and evaluating design schemes based on the designer’s subjective experience in the original design process and provides the scientific design method for integral form design of large-scale mechanical and electrical products.

关 键 词：感性工学 BP神经网络 程控切纸机 整体造型 感性评价矩阵 

分 类 号：TB472[一般工业技术—工业设计]