Research On VBHF Technology Of Aluminium Alloy Stamping

Recently, aluminum alloy sheet and other lightweight materials have been applied in automobile industry more and more popularly, which will decrease the auto body weight greatly and thus reduce gas consumption and emission. However, the formability of these materials are not as good as traditional automotive steel sheet, and how to improve the forming result becomes one of the most popular and important problems. As a simple and effective technical solution, Variable blank holder force (VBHF) technology has attracted more and more people to research on it. Hemispheric part is a basic and typical surface shape part. Research on its deforming law and characteristic is very important to get better forming quality and reveal the drawing mechanism of surface shape part. Variable blank holder force which changes with the deep drawing time can improve the forming quality, but the relationship between the forming quality and the values at different time is not clear. The development of artificial intelligence technology is a possible study method to get an optimized variable BHF.In the paper, based on numerical simulation software, experiment of different variable BHF change with the deep drawing time was conducted. Intelligent optimization method was then used to figure out the influence of BHF values on forming quality. The main work of this thesis can be concluded as follows:A standard was proposed to evaluate the forming quality, so we can evaluate the forming quality is good or bad by compare the value of number. The standard was applied to process the orthogonal test samples, and the influence of BHF combination on forming quality is obtained by means of range analysis. The sequence of influence extent is:F4,F3,F1,F2, F5(F1-F5are BHF at t=0,1/4T,1/2T,3/4T, T. T is the time of deep drawing). Meanwhile, the optimum BHF combination is got, and the corresponding forming quality indicator by CAE simulation confirmation test is0.0997, which is the least among all the orthogonal samples.Artificial Neural Network(ANN) model is developed to map the complex non-linear relationship between forming quality indicator and BHF values, and Genetic Algorithm (GA) is used to optimize BHF value in the selected range of parameters, and the best variable BHF curve is figured out. The forming quality indicator of CAE verification test is0.0595, which is less than all the other test samples. The result of physical deep drawing experiment of5083aluminum alloy indicats that the quality of part formed with the optimum variable BHF curve is better than other variable BHF, so the proposed method using ANN combined with GA to optimize the variable BHF curve is feasible.