【摘要】 冲压过程中,因材料的性能参数、润滑条件的波动和模具参数的改变,导致冲压件质量的不稳定,因此必须提高冲压件质量的控制鲁棒性。压边力是板料冲压过程的重要工艺参数之一,合理控制压边力的大小,可以避免起皱或破裂等缺陷。文章提出了一种基于结合RBF神经网络和经典状态空间理论的板料成形过程控制策略,并建立了其控制模型。引入时序分析法,将压边力的调整与反映工件成形质量的法兰边的吸入量误差联系起来,同时应用非线性最小二乘法识别出增益系数。仿真结果表明,在加入一噪声后,仍能获得满意的冲压件质量,模型具有较高的控制鲁棒性。更多还原

【Abstract】 It will lead to instability of the stamping quality during the forming process because of randomized variation of known process,e.g.material parameters,geometric variations,lubrication(friction),etc.,and it will affect the stamping quality in the end,so the control robustness of stamping quality must be improved.Blank-holder force(BHF) is one of the most important process parameters in the stamping process,and it will avoid the defect of stamping such as wrinkle and fracture by controlling the reasonable BHF.A new control strategy is proposed based on the classical state space control theory combining with RBF neural network during the sheet metal forming,and the mathematical model of process control is set up.The relation between modifying of BHF and error of the blank draw-in is built by introducing time series analysis,and the gain factors are identified by using non-linear least square solver.The simulation results indicate that it still gains the good stamping quality by using the proposed control strategy when it has the randomized noise of known process during the forming process,and the model has a high robustness.更多还原