【作者】 张书涛；

【导师】 胡赤兵； 苏建宁；

【作者基本信息】 兰州理工大学 ， 机械制造及其自动化， 2014， 博士

【摘要】 产品造型形态是表达产品设计思想与实现产品功能的语言和媒介,能够传达精神、文化等层面的意义与象征性。它是产品自身功能、结构、材料以及工艺技术等客观因素与设计师和用户在审美、价值判断等主观因素相互作用的综合结果。基于此,本研究针对产品意象造型智能设计相关问题,依据用户与设计师的认知思维规律,寻求进一步提升产品意象造型智能设计系统性能的方法以及设计思维品质评价的方法。本研究的主要研究内容与获得的结论如下。(1)提出应用数量化理论建立符合人的视知觉选择分辨性和简约调节性的产品意象造型设计要素辨识模型。应用该模型在感性工学系统的理论架构内对产品造型设计要素进行分解,以造型设计要素为自变量、用户情感需求为因变量,探讨用户的感性意象与产品造型设计要素的对应关系,并据此辨识产品造型设计要素针对各感性意象的个性要素和平台要素。建立计算机辅助产品意象造型设计要素分析系统,将该系统所分析的产品造型设计要素数据应用于实际项目,取得了较好的效果。(2)依据人的视知觉整体组织性、恒常记忆性、选择分辨性和简约调节性构建神经网络,提出了改良型四层BP神经网络。基于改良型四层BP神经网络建立以产品形态要素为输入、用户情感意象需求为输出、用户情感需求调查结果为训练样本的产品意象智能评价系统。与基于常规三层BP神经网络的产品意象智能评价系统进行对比研究,结果表明基于改良型四层BP神经网络的产品意象智能评价系统具有较高的预测精度。(3)通过分析设计师的设计思维流程及其设计进程、设计思维模式与设计策略所构成的关系单元,提出了基于广义交互式遗传算法的设计师设计思维模型。应用基于精英保留策略的非劣分类遗传算法建立人机交互式产品多意象造型智能设计系统(IIDSPMIF)模拟产品设计过程中的多意象目标优化。实验结果表明,IIDSPMIF系统总体上符合设计师的设计流程,能够模拟设计师的设计策略和设计思维模式。(4)提出设计思维品质评价方法,该方法应用思维数学的相关理论将产品设计过程转换为设计师的思维网络,采用系统套、思维坐标系等理论量化思维环节,求解设计思维品质函数。应用该方法分析了设计师之间的思维品质差异以验证其可行性。最后,应用该方法对比分析了广义设计师C-IIDSPMIF与设计师C的思维品质,结果表明IIDSPMIF系统有效提升了广义设计师的思维独创度、统摄度、流畅度等方面的思维品质。本研究所取得的研究成果为指导企业开展新产品的意象造型设计进行了有益的探索与尝试。另外,设计思维品质评价方法的建立为研究设计认知思维提供了新的实践与理论基础。更多还原

【Abstract】 Product form is the expression of design ideas and the language and media of product features and conveys the meaning and symbolic of spiritual, cultural and other aspects. It is the interaction result of the objective factors of product function, structure, materials, technology and the designers’and users’subjective factors of aesthetic and value judgments. Based on this, the study analyzed the design issues related to the intelligent design of product image form and took the visual perception characteristics as a starting point to seek the method which can enhance the capability of the product image form intelligent design system and the design thinking quality evaluation method based on the law of cognition of users and designers. The main research contents and conclusions are as follows.(1) The study established the product image form factors identification model that is according to the simple regulating and identifiable discrimination of visual perception with quantification theory. Taking product form factors as independent variables and the emotional needs of users as dependent variable, the model was used to explore the relationship between users’Kansei image and product form factors with Kansei Engineering theory. And based on this, the individual factors and platform factors of product form to each Kansei image were identified. A computer aided analysis system of product image form factors was built. The analysis data was used in actual project and achieved good results.(2) According to the overall organization, constant memory, simple regulating and identifiable discrimination of visual perception, an improved4-layer BP neural network was established. Taking product form factors as input data, users’emotional needs as output data and the survey results of users’emotional needs as training samples, a product image intelligent evaluation system was built based on the improved4-layer BP neural network. A comparative study with the product image intelligent evaluation system that is based on conventional3-layer BP neural network was taken. And the results showed that the product form image intelligent evaluation systems based on the improved4-layer BP neural network is with high prediction accuracy.(3) Based on the analysis of design thinking process and the relationship unit of design process, design thinking mode and design strategy, a design thinking model was built with the generalized interactive genetic algorithm. An interactive intelligent design system of product multiple image form (IIDSPMIF) was established to simulate the multi-image-objective optimization in the design process. The experimental results show that IIDSPMIF system meets the design process generally and simulates the design strategy and design thinking mode well.(4) The study proposed a new design thinking quality evaluation method that translates the product design process into thinking network and quantifies thinking links with system sets and thinking coordinates to calculate the function values of design thinking quality. The method was used to analyze the differences of designers’ thinking quality to verify its feasibility. Finally, the method was used to analyze the thinking quality of designer C and the thinking quality of the generalized designer C-IIDSPMIF. The results showed that the IIDSPMIF system effectively enhances the creation degree, gathering degree, easy-smooth degree and other aspects of the generalized designer’s thinking quality.The research results obtained in this study is a beneficial exploration and attempt to the image form design of new product. In addition, the design thinking quality evaluation method provides a new practical and theoretical foundation for the study of design thinking.更多还原