Research on cute semantic CMF design based on product digital twin principle

Abstract Based on the principle of digital twinning and Kansei Engineering, this paper analyzes the product modeling data to obtain the product lovely Kansei semantic reference data based on personal aesthetic needs. Firstly, the twin data of product family is collected and extracted as the training set, and then the k-Means clustering algorithm is used to cluster the twin data, and the training results are converged to obtain the virtual model database. Then, the twin data which need to be improved are compared with the data in the model database, and adjusted flexibly. According to the guidance of the data, the appearance features which are subjectively similar to the perceptual product semantics of the training system are obtained. At present, the system realizes the simulation model and algorithm of color, material and surface processing ( CMF ) twin data under the cute perceptual semantics, and constitutes a model database for ' cute-malicious ' phrase pairs under the perceptual semantic vocabulary library. Using the twin data analysis results of the same style product to assist product appearance design can optimize the design scheme and simplify the design complexity.