Research On Product Design Method Driven By User Style Image Cognition

With the rapid development of information acquisition technology,manufacturing technology and automation technology,the design and manufacturing ability of various enterprises on the basis of user's choice of elements such as the function and quality of the same type of products is becoming more and more perfect.However,this leads to the situation that a large number of products are gradually equal in the level of function and quality,which is difficult for users to choose.Facing such a market environment,product design is always the key point for enterprises to stand out in the fierce market competition.As the key content of product design,the innovation of product modeling design has naturally become an important way for enterprises to improve the added value of products and win the choice of consumers.Often successful product modeling design can not only make the product attract the attention of consumers at the first time,meet their personalized psychological demands for the product,but also make the product have certain brand and cultural connotation value,stabilize the user group of the product,and obtain the support of the target market and consumers,which also reflects the importance of product modeling design research.Generally speaking,product modeling design is realized through the design combination of product modeling outline,color,material texture and decorative pattern.However,in the early stage of modeling design,to realize the concept of rapid modeling design,it is mainly for the design exploration of the shape features(mainly the shape features of the direction,line and surface)in modeling design The focus of the research is the innovative design method and technology for the early form characteristics of product modeling design.The research of product modeling design involves many theoretical methods,such as knowledge processing method,design cognition,data mining technology,design psychology,Kansei Engineering,computer-aided product modeling design technology and so on.Based on the cognitive mechanism of consumers to product modeling,this paper studies the influence mechanism of modeling influencing feature elements(point,line,surface)and product style cognitive image on product modeling design results according to the transfer and reuse process of knowledge in the design process among different design subjects,so as to solve the problem of product modeling for personalized user cognitive image demand Finally,based on the theories and technologies of cognitive psychology,pattern recognition,Kansei Engineering,computer-aided design technology,shape grammar,product modeling innovation design method,a set of product modeling generation model for user's personalized needs is established Innovative design methods,and develop the corresponding aided design prototype system.The main research contents are as follows:1.Research on method architecture of product intelligent design driven by user style image cognitionDue to the coexistence of sensibility and rationality,qualitative and quantitative characteristics of user image cognition caused by product modeling design,it is difficult for traditional knowledge processing mode to complete the construction of knowledge processing model in the field of product modeling design comprehensively and effectively,thus the product modeling design method based on knowledge reuse cannot be realized.Aiming at the above problems,based on the analysis of the generation mechanism and function of style image in the field of product modeling design,the knowledge acquisition in the field of product modeling design,and the design method of product modeling knowledge reuse based on user image cognition,the intelligent design technology is applied as the support,and the method architecture of product intelligent design driven by user style image cognition is constructed According to the key technology,the corresponding research ideas are put forward.2.Research on semantic space construction of product cognitive image based on Data MiningThe traditional modeling design method is difficult to deal with the fuzziness and uncertainty of the user's style image demand,and can not provide effective means for designers to solve the problem that the design can not meet the user's demand caused by the subjective deviation of style image cognition between users and designers.This paper studies the semantic acquisition and representation of user's cognitive image based on data mining,obtains the user's comments on the product through web crawler technology,and after data preprocessing,combined with the quantitative technology of perceptual image in Kansei Engineering,proposes the semantic quantitative method of user's cognitive image based on emotional dimension word frequency method(tf-ed),and constructs the user's product cognitive image Semantic space provides data support for the semantic association model of modeling parameters and user cognitive images.3.Research on product modeling generation design method based on Parametric Design TechnologyAiming at the problem that the commonly used parametric representation methods of feature points,lines and surfaces are difficult for designers to quickly reuse design due to the large amount of parameters in complex modeling,and the traditional shape grammar is applied to product modeling design because of the inconsistency of initial shape and rule details,which makes the generation design effect based on feature rules not ideal,and the reusability is poor.This paper presents a method of product modeling generation design based on improved parametric shape grammar(PSG).Firstly,the parameterized representation of modeling features based on MBB algorithm is studied,which unifies the expression form of complex modeling features and facilitates the reuse and innovation of composition rules.Secondly,based on the parameterized shape grammar described by improved rules,the basic rules of grammar and rule coding are studied,and the generation design method based on rule reuse is proposed.Compared with the commonly used parametric design method of modeling features,this method can help designers directly adjust the design parameters and rules,quickly complete the design,and improve the design efficiency.4.Research on product modeling optimization design method based on Intelligent AlgorithmAiming at the prediction and evaluation of product modeling style image semantics,the nonlinear mapping relationship between product modeling design element parameter set and user's cognitive image semantics is established by using artificial neural network(ANN),and the prediction network model of product modeling style image semantics is constructed.Taking the predicted value of artificial neural network as the fitness value in genetic iteration,and according to the global optimization characteristics of genetic algorithm(GA),the problem of product modeling optimization design is solved.Finally,through the actual case,it is verified that the model can not only accurately predict the customer cognitive image value caused by modeling,but also calculate the optimal product modeling design parameter solution in line with the image value after combining with the genetic algorithm.The optimal parameter solution can generate the most consistent consumer modeling style image cognition through reverse decoding according to the proposed parametric product modeling generation design method The demand of product modeling design.5.Development of product intelligent design(USIC-DIPD)prototype system driven by user style image cognitionBased on the research of the key technologies such as the construction of user style image cognitive space based on data mining,product modeling generation design method based on parametric technology,and product modeling optimization design method based on intelligent algorithm,a product intelligent design prototype system driven by user style image cognition is developed for designers and users,which provides a reference for product design based on knowledge reuse The new design provides the corresponding theoretical and technical research reference.