Research On The Sensory Feeling Of Product Design For Electric Toothbrush Based On Kansei Engineering And Artificial Neural Networks

Nowadays,consumers pay more and more attention to the emotional image of product modeling,and electric toothbrush,as a new practical household cleaning tool in recent years,is widely favored by people.Therefore,this study explores and analyzes the correlation between product design elements and users’ emotional images by mining the shape design features of electric toothbrush and users’ emotional images,and applies guidance to the optimization and development of the shape design indicators of electric toothbrush,so as to design electric toothbrush products in line with their emotional images for consumers.This study uses web crawler technology to collect online product reviews(215,827items)and samples(200 models)from consumers;And then the huge reviews and sample sets were extracted through Word2 vec,factor analysis,multi-dimensional scaling and clustering methodology,and 3 groups of representative Kansei words and32 representative samples were obtained.The semantic differential method was used to evaluate consumers’ Kansei images,and a total of 415 valid samples were taken.Using the semantic difference method,a questionnaire was used to evaluate the perceptual images of consumers,with a total of 415 valid samples taken.Finally,the linear prediction model of the relationship between product design elements and consumer perceptual images and the non-linear prediction model of Back-Propagation Neural Network(BPNN)and Genetic Algorithm-based BPNN(GA-BPNN)are constructed by using the Quantification Theory Type Ⅰ(QTTⅠ).The average error discrimination method was used to compare and evaluate the three models,and the results showed that the prediction effect of QTTⅠ model was more accurate.Then 5 designers were invited to conduct product design and evaluation verification through independent design and design indexes generated according to the optimal model.135 of the five groups of samples were used to conduct the paired samples T-test,and the results showed that the p values were all greater than 0.05,indicating that the linear prediction model of QTTⅠ in this study was reliable and valuable in application.Moreover,it can effectively provide the correlation table of the influence of modeling items and categories on perceptual image,which can be used as a guide for the design index of product modeling image in this study.Finally,based on the optimal design index and the guiding principles of ergonomics and aesthetic concepts,the practical design and development of the product were carried out,and the integrated design of the shape,color and material,man-machine size design,internal structure design and man-machine interaction interface design of the electric toothbrush was completed.Finally,the product modeling was completed and the design effect drawing was generated.Based on web crawler technology and systematic analysis,this study proposes a set of scientific and complete optimization design scheme of product perceptual design,which can provide clear design index and reference for designers’ traditional black box design mode.