基于产品外观形态解构的感性意象模型研究

发布时间：2018-11-18 14:27

【摘要】：如今消费者逐渐重视对感性的需求，感性意象的定量研究已成为当今设计学科热点研究的问题之一。感性工学为设计师提供了了解消费者内心感受的重要工具。本研究将运用感性工学研究方法，以智能手机为例对其外观形态进行解构分析，研究产品造型设计的类目与项目，进而探究造型设计要素的感性特征及其感性意象空间的关联性。在大量调研与分析的基础上，筛选目标产品的研究样本，，运用形态分析法，对产品外观形态进行解构，并确定产品造型要素及其项目，从而获得样本吸引消费者的关键影响因素。结合正交试验的方法，将多个不同设计元素或构件进行排列、替换，组合变化产生新的造型，以确定典型代表实验样本。搜集感性意象语汇，建立资料库，并利用多元尺度、聚类分析等方法，筛选并确定代表性感性意象的语汇；结合产品造型样本，进行感性评价试验，运用语义差分法将人们对于产品外观形态的主观评价转化为量化评价值，从而确定消费者的感性需求。基于产品形态要素量化值与产品外观形态主观评价量化值，进行基于数量化I类分析方法的产品造型意多元线性回归模型构建。并以此预测自变量与因变量之间的影响程度，建立关于感性意象的产品造型要素设计法则；对所建立的模型进行验证，利用统计学检验方法来分析模型的可靠性。鉴于线性模型的局限性，基于非线性BP人工神经网络理论，结合产品形态要素与感性意象评价的量化值，确定多层神经网络结构，训练网络达到预定目标。构建关于产品形态要素的感性意象预测模型。通过比较误差评价所建立的产品形态法则，得到不同建模方法的优劣性。并利用造型意象非线性模型预测结果，分析造型要素对感性意象影响规律。
[Abstract]:Nowadays, consumers pay more and more attention to the demand of sensibility, and the quantitative study of perceptual image has become one of the hot issues in the field of design. Perceptual engineering provides an important tool for designers to understand consumers' inner feelings. This research will use the perceptual engineering research method, take the smart phone as an example to carry on the deconstruction analysis to its appearance form, studies the product modelling design category and the item, Then explore the emotional characteristics of the design elements and the relevance of perceptual image space. On the basis of a great deal of investigation and analysis, the research samples of the target products are screened, the appearance of the products is deconstructed by morphological analysis, and the elements of product modeling and their items are determined. Thus obtaining samples to attract consumers is a key factor. Combined with the orthogonal test method, several different design elements or components are arranged, replaced and combined to produce a new shape to determine the typical representative experimental samples. Collect the words of perceptual image, set up the database, and use the methods of multivariate scale and cluster analysis to screen and determine the vocabulary of representative perceptual image; Based on the sample of product modeling, the perceptual evaluation test is carried out, and the subjective evaluation of product appearance is transformed into quantitative evaluation value by using semantic difference method, so as to determine the perceptual needs of consumers. Based on the quantitative value of product form elements and subjective evaluation value of product appearance, the multi-linear regression model of product modeling meaning is constructed based on quantitative class I analysis method. The influence degree between independent variable and dependent variable is predicted, the design rule of product modeling elements about perceptual image is established, the established model is verified, and the reliability of the model is analyzed by statistical test method. In view of the limitation of the linear model, based on the theory of nonlinear BP artificial neural network, combined with the quantitative value of product morphological elements and perceptual image evaluation, the multi-layer neural network structure is determined and the training network achieves the predetermined goal. The perceptual image prediction model about product form elements is constructed. The advantages and disadvantages of different modeling methods are obtained by comparing the product morphology rules established by error evaluation. The influence of modeling elements on perceptual image is analyzed by using the prediction results of modeling image nonlinear model.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TB472

【参考文献】