Font Size: a A A

Research On FLSD And Application In Multistage Sheet Metal Drawing

Posted on:2007-06-20Degree:MasterType:Thesis
Country:ChinaCandidate:X WangFull Text:PDF
GTID:2121360185959803Subject:Materials Processing Engineering
Abstract/Summary:PDF Full Text Request
The strain-based forming limit diagram is generally obtained on the linear (proportional) or near linear loading condition. So it often leads to erroneous assessments on the analysis of blank sheet formability under complex strain paths by the FLD criterion. In recent years, most researchers consider that the ultimate strain state is determined only by the final stress state, the stress state of the instability zone is independent of strain path, and propose the stress-based forming limit diagram (FLSD) criterion. The obstacle of the research of FLSD and its application lies in the determination and validation in engineering.In this thesis such work was conducted: obtaining the limit strain of two ferrous metals (SPCEN,08Al) and one non-ferrous metal (LF21M) under complex strain paths by experiments, and gaining limit stress based on the stress-strain relationship program, which developed by the member of the project team; furthermore, in view of the nonlinear correlation between the FLSD and the forming condition (stress state), performed the research on the prediction of FLSD by the BP artificial neural network because of its high nonlinear mapping function; carrying out the finite element simulation of the multistage deep drawing of the left rear suspension bracket of the Palio car by Dynaform, analyzed the formability of the parts undergoing multistage deep drawing according to the stress-based forming limit diagram from experiments and prediction.The results show that the FLSDs obtained under the different loading paths are approximate to one curve, which means that the stress-based forming limit diagram is independent of strain paths. It is feasible and effective to analyze the sheet metal formability utilizing the FLSD as a criterion, and to predict the FLSD through artificial neural network (ANN). The thesis validates through actual experiments that FLSD is much more reliable and effective than the strain-based FLD in determining the multistage part formability under complex strain paths.
Keywords/Search Tags:formability, FLSD, finite element simulation, strain path, artificial neural network, multistage deep drawing
PDF Full Text Request
Related items