高强度金属板多道次渐进折弯成形及回弹研究-材料加工工程专业论文.docxVIP

华 华 中 科 技 大 学 博 士 学 位 论 文 IV IV air-bending forming field. Based on the tests for air-bending of sheet metal, a prediction model of springback radius and punch radius is developed with GA(genetic algorithm) and BPNN(back propagation neural network). BPNN is prone to getting into local extrema and convergence is slow. To overcome these drawbacks, this study attempts to combine GA, avoiding local minima and achieving global convergence quickly and correctly by searching in several regions simultaneously, with BPNN to minimize the total mean squared error (MSE) between actual output of the network and the desired output through optimizing the weights of neural network. A mathematical model of springback is developed with dimensional analysis and orthogonal test. Dimensional analysis is used to confirm the relationship of the physical variables with the principle of dimensional homogeneity in physical laws and the Buckingham pi law. And the corresponding relationship between the sheet metal springback radius and the sheet thickness, yield strength ,the punch radius ,the Young’s module could be solved quantitatively. A model of springback correction is established with off-line closed-loop control system. Introducing feedback control idea in automatic control theory to multiple-step incremental air-bending forming of sheet metal, an off-line closed-loop control iterative algorithm, combined by Fast Fourier and wavelet transform, is developed. With this algorithm, the mould surface of sheet metal incremental air-bending forming could be properly corrected, and the springback errors of the formed workpieces could be effectively reduced. Combining the above three models with numerical simulation, which can be taken as a new approach for sheet metal multiple-step incremental air-bending forming, tool design and warpage deformation analysis. Based on Hill’s yielding criterion and exponential strain hardening law as well as plane strain condition, ABAQUS 3-D finite-element m