Research On Key Techniques For Digital Development Of Auto-panel Geared To Mass Customization

Today's world market, teemed with individualized needs, is dynamic and constantly changing. One of the most distinctive features of modern manufacturing is that it should meet customers' individualized needs, and respond quickly to market demands. Thus mass customization is becoming the developing trend for modern enterprises if they want to seek and maintain their competitive edge. China has witnessed the formation of the second biggest auto market in the world, with the individualized purchasing needs increasing sharply. During the process of research and development of new products, what weighs most among the total cost of auto development is that of the auto-body product and of its processing equipment. The design and manufacturing of automotive dies is vital in that it affects the efficiency and quality of auto-making. Therefore, the technological innovation of designing the auto-body product is of practical significance to increasing the manufacturing capability of China's auto industry.The success of mass customization can be assured with the appropriate auto-panel product family planning and the optimum level of product family customization. This dissertation studies the connections between product customization level and customization cost, between the level and customer value. By analyzing the disparity between customized product and individual customer's expectation, it advances a quantitative solution to enterprises' customization degree and the method of determining its optimization. Similarly, based on an analysis of the customer value of customized product, the research paper puts forward a quantitative solution to clients' customization level and the approach to determining its optimization. The double value of enterprise and market provides sufficient ground on which the optimal strategy of choosing product family customization level can be devised. The implementation of the strategy will not only make the automotive enterprise more profitable, but also satisfy customers' individualized demands better.With the design of the forming-simulation of auto-body covering part as a starting point, the present article deals with the digital modeling based on mass Customization. Feature-modeling method of auto-panel, whose curving and modeling features have been defined, is her studied. The paper puts forward the multiple-factor coupling simulation of stamp-forming for auto panel with the practice test. By studying the factors influencing the resilience of sheet metal and the approach to the reduction of spring-back, it clarifies the thought of optimizing the stamping process planning for auto panels. Based on digital-analysis of panel product, it summarizes the technologies to improve the quality of sheet metal forming, and precautions against the invalid sheet metal forming.The purpose of digitalized product analysis is to optimize product design and guarantee product quality. With a proper size of the blank for auto panel, the formability and quality of the componential parts can be enhanced. The present paper studies the practicability of soft computing and its combination and integration. By drawing upon such genetic-neural networks as ANN, feed-forward neural network, BP algorithm, and IGA to forecast the size and form of auto panel, it illustrates how to test the forecasting accuracy of IGA, BP algorithm, and FBP, thus coming up with the neural networks suitable to the foretelling of auto panel's sizes and shapes.The digitalization of auto-body covering product manufacturing is heralded by that of die-making. The prevalent CAD/CAM software systems attach much importance to the information integration and management concerning the processing of complex curved surfaces, and are furnished with such functions as curve and surface fitting, and their interpolation. The quality and efficiency of die-face processing can be enhanced if the NURBS interpolation technology is applied in the NC processing of auto panel's stamping die-face to determine the cutter's path. The present dissertation provides the geometry modeling of surface scattered data by NURBS. In the original designed surface are randomly distributed huge amount of scattered data, which then are clustered by the neural network based on SOFM, thus the NURBS curve is constructed, and interpolation surface reconstruction is completed. Besides, through simulated tests, the paper conducts an error analysis concerning the original and the reconstructed surface.Now ERP has been the essence of enterprise information building, the integration of ERP and PDM is going to be the developing trend for enterprises. Considering the demand of digitalizing the development of auto-body covering part, based on an analysis of its assorted information and data, the dissertation presents a PDM solution to the mass customization of auto panel, hence a hierarchical collaborative application of the panel product and its derivatives. By digitally simulating product design and manufacturing, it constructs the integrative framework of the structure and application of the product data management prototype system, thus realizing the adaptability and compatibility of the PDM prototype.The final section of this dissertation reiterates what the author has achieved in his research, and predicts the possible developing trends, both theoretically and technologically, in this research field.