Research On Humanized Design Method For Earphone With Surface Form~*

With the rapid development of science and technology,people’s living standards have been improved day by day,and users pay more attention to the comfort and amenity of products.Considering the comfort of wearable products with direct contact with human body has become a hot spot in current design research.This paper presents a humanized design method of earphones with surface modeling,which takes the conchal surface as input,and takes the reconstruction,clustering and recognition of the characteristic surface as technical means to balance the user’s comfort demand and the market’s batch production demand.The main research work is as follows:(1)A method for measure and analyze the characteristic dimensions of the auricular concha is established by automatically and accurately extracting the key characteristic points of auricular concha from collected samples.According to the 3D scanning technology,the 3D models of outer ear of 315 young Chinese male and female aged 18-28were collected.The 11 key characteristic points of auricular concha are defined by the domestic and foreign medical research literature.To overcome the shortcomings of traditional anthropometry technology,an automatic and accurate extraction method of 3D coordinates of conchal key characteristic points is proposed based on the curvature principle of NURBS surface.Through the method of mathematical statistical analysis,the bilateral differences and gender differences of young Chinese auricular concha were systematically analyzed,and then the individual,gender and racial differences of auricle conchae were pointed out,as well as the necessity of constructing the database of Chinese ear morphology.(2)A "bidirectional first-order contour reconstruction" method for obtaining complex surface data points is proposed and applied to the surface reconstruction of the concha.In order to describe the triangular mesh surface of different samples with the same number and identical data points on the respective surfaces,a "bidirectional first-order contour reconstruction" method is proposed.The reconstruction surface of the auricular concha with the same topological structure is calculated based on the interpolation method of NURBS surface by taking 795 data points as the data points.Then the accuracy,continuity and smoothness of reconstruction surface are tested by means of surface deviation analysis,curvature analysis and texture analysis of curved zebras,and the feasibility of reconstruction method of conchal surface based on interpolation of NURB S surface is demonstrated.(3)An improved hierarchical clustering algorithm for conchal surface morphology was established,and the conchal surface morphology database for young Chinese people was constructed for earphone design.To overcome the shortcomings of traditional hierarchical clustering algorithm,an improved hierarchical clustering algorithm based on the shape classification of conchal surface was constructed,and the criterion of the best clustering group was introduced into the algorithm.According to the improved hierarchical clustering algorithm,the conchae of young Chinese male and female aged 18-28 are divided into 29 categories,and the advantages of the improved algorithm are demonstrated through comparison with the results of the traditional hierarchical clustering algorithm.Based on the interpolation method of NURBS surface,the conchal common characteristic surface for each cluster is calculated.The reliability for the result of the improved hierarchical clustering algorithm is verified by the deviation analysis between intra-group and inter-group samples.(4)The evaluation method of humanized design results for earphone is established.According to the constructed database of the conchal morphology,the earphone is designed and 3D printed.Through the wearing and motion test,the anti-slip of the earphone and the users’ subjective comfort are tested.Then the objective evaluation method of the wearing comfort of the earphone based on the finite element simulation analysis of the outer ear-earphone is established.Finally the validity of the humanized design method constructed in this paper is verified by the evaluation results.(5)A customized design method for earphone is proposed.The shape recognition model of the conchal surface based on K-nearest neighbor(KNN)algorithm and probabilistic neural network(PNN)algorithm is constructed respectively.The conclusion is drawn that the PNN model has a high accuracy rate for the shape recognition of concha is obtained.The earphone customized design method which combines the PNN recognition model of the conchal morphology database is determined.Finally the customized design example verifies the feasibility of the method.(6)According to the above theories and methods,the Rhino-Script,the scientific computing language MATLAB,the statistical analysis software SPSS and the finite element analysis software ABAQUS are used to extract the conchal surface data points,clustering,recognition calculation,data statistics analysis,and outer ear-earphone contact stress simulation analysis,respectively.Meanwhile,the RHINOCEROS and CATIA were used to process and analyze the deviation between each conchal surface.Finally,a large number of data and charts were obtained,which provided a basis for the design of ear-related products.The research results of this paper have theoretical significance for the humanized design of most wearable products with surface modeling,and have important practical application value for guiding enterprise R&D and production practice.