Research On The Assistant System Of The High Speed Blender Form Design Based On Kansei Imagery

With the upgrading of consumer demand,the practical functions such as product structure and function are no longer the core influence of consumer behavior.The deeper individualized and fashionable emotional has become the new pursuit of consumers.Small household appliances are products that are closely related to people's daily lives.Consumers have higher Kansei demand for their personalization and humanization.In order to design a kansei product that meets consumers' expectations,designers will analyze the relationship of consumer kansei needs and product design.The high speed blender is researched by the thesis.Guided by the theory of kansei engineering,quantification theory type I and BP neural network are used to explore the correlation rule between the consumer's kansei demand and the form of the high speed blender.Based on this,VB and Access were used to construct a kansei image design assistant system,which provided a more intelligent reference for designers in the early stage of product design and in late period of market positioning.The main research work and achievements are as follows:(1)Using focus group method,KJ method,MDS,K-means cluster analysis,the sample is subjected to dimensionality reduction clustering,and 25 representative samples are selected.(2)Using the morphological analysis to analyze and extract the form elements of the high speed blender,and the form element matrix of the representative sample is encoded by two coding methods.Providing the form evaluation standard for the design assistant system;.(3)Using the semantic difference method,the test-retest reliability method,factor analysis,and hierarchical cluster analysis to investigate the kansei image of the high speed blender,extracting five pairs of kansei imagery,and Quantifying the kansei image of consumers on the form of the high speed blender.(4)Using the quantification theory type I to construct the linear correlation model between the form elements and the kansei image of the high speed blender;The BP neural network about the form elements and kansei images of the high speed blender was trained by Matlab to analyze the adaptability of the linear correlation model and the nonlinear correlation model for the experiment.based on the verification sample,the fitness test of the nonlinear correlation model and the linear correlation model is carried out,and the linear model is found to have a higher fitness,which is used as prediction mechanism for the sample and reasoning mechanism in the design assistant system.(5)Based on the relevance research,the VB is combined with the Access to construct a high speed blender form design assistant system based on kansei imagery.In this study,the image design and image design assistant system of the high speed blender were explored.The theoretical research of kansei engineering was combined with computer aided technology and artificial intelligence technology to solve the correspondence between form and kansei image with a more intelligent method.The system has enabled kansei image obtain an efficient and intuitive response,and at the same time,the designer's case has been effectively evaluated and analyzed.This system reduces the blindness of the design,reduces the design cost,shortens the design cycle,and improves the design fault tolerance.