Electromechanical Product Modeling Image Prediction Model Research Based On BP Neural Network

With the consumption concept of persons changing,product design concept changes from the "product priority" to "consumer demand for the spirit ".Consumers describe the image of the product with ambiguity,uncertainty,so designers must design the product elements with the perception of consumers linked in order to meet consumers needs.In this paper,we use kansei engineering to quantify the consumer’s product modeling quality,and use BP neural network to construct the relationship between product modeling and emotional image.Based on this,we develop the the system of forecasting product modelling emotional image.Designers grasp consumer demand for products improving the effectiveness of product design.Firstly,we analyze the relationship between electromechanical product modeling and emotional image.Taking the car as an example,the image vocabulary and modeling samples of electromechanical product were collected through the design investigation,carrying out statistical analysis method,semantic difference,triangular fuzzy number.We determine representative image vocabulary of describing electromechanical product and typical product modeling by using SPSS,multidimensional scale analysis,factor analysis and cluster analysis.Secondly,through the analysis of the modeling elements of electromechanical products(cars),the types of geometrical shapes,colors,materials and the overall characteristic elements are selected,and the elements that influence image of the electromechanical products(cars)are selected.According to the information entropy and the analytic hierarchy process(AHP),respectively,we determine the weight of the tested persons and product modeling element.Thirdly,BP neural network is used to construct the relationship between the modeling elements and the emotional image of the electromechanical product(car),and the MATLAB and the sample data are used to train the emotional image prediction model of the product design scheme.Finally,the application of MATLAB user graphical interface(GUI),access database technology develop electromechanical products(car)modeling customer modelling image prediction system prototype.