Research On Springback Evaluation For Automotive Panels And Inverse Determination Of Material Parameters

Automotive panels that usually characterized with large size, complexstructure, numerous freeform surface and high surface quality requirements.During the stamping forming processes, it is prone to appear the springbackdefects after unloading, which was not only seriously impacts on the stampingsheet metal forming quality and welding assembly process accuracy, butgreatly increse the difficulty automotive panels of automotive panels moulddesign difficulty. Therefore, the accurate prediction of the springback andhow to control it had become an essential problem to be solved in the currentsheet metal forming process. Springback is impossible completely to beavoided. However, improving the accuracy of predict springback andcontrolling springback had become the key subject that must be settled in thecurrent stamping field. If we can acquire the exact springback deviation oneach surface of stamping a products part at the stage of mould before thedesign of mould, defects caused by springback can be reduced through theinverse compensation trategy. The key lies in the accurate prediction ofspringback occurring position and accurate measurement of the amount of it.At present, finite element simulation technology is widely used to predictthe springback in sheet metal forming. This paper expounds the basic theoryof sheet metal forming simulation technology, and introduces some commonlyused professional simulation software. Some key factors that may result tospringback simulation error are analysed systematically.Due to the various structures of the automotive panel and the variation ofmaterial property in actual stamping process, accurate prediction ofspringback through simulation technology becomes more and more difficult.In this paper, the typical springback characteristics are analyzed. For one typeof part structure, determination of its sensitive parameters mostly attributed tothe extent of springback is made. It is found that the elastic modulus coefficient of strain hardening and hardening exponent have significantimpact on the amount of springback.A genetic optimal algorithm combinedwith simulation technology is adopted to revise the material performanceparameters. It demonstrates that the revised material parameters can moreaccurately describes the real performance in sheet metal forming.Considering that enterprises are not convenient to obtain metal propertiesdata, a shared database of material performance parameters is establishedbased on Web. This database works stably and correctly.In order to realize the measurement the springback in3-dimensions oflarge-scale complex automobile panels, this paper presents a method based onbinocular stereo vision technology to fulfil the evaluation extent ofspringback. It is carried out by matching the cloud data model,which isobtained from stereo vision technology, with the original CAD model inoptimal position and calculate their normal deviation, to realize the digitalmeasurement of2D cross section, three-dimensional surface and the specificlocation of a part. The digital measurement results are used to compare withthat of numerlcal simulation using the revised material parameters. The resultsshow that it can significantly improve the accuracy of springback predictionby using the revised material parameters for springback simulation.