车辆底板冲压的响应面拟合与改进蜂群算法优化

Response surface fitting and improved bee colony algorithm optimization for vehicle bottom plate in stamping

为了减小车辆底板冲压件的最大减薄率,提出了基于改进蜂群算法的工艺参数优化方法。选择压边力、冲压速度、模具间隙为优化参数,依据最优拉丁超立方抽样法设计了实验,使用Dynaform有限元软件得到了实验仿真结果。建立了质量参数与工艺参数的响应面模型,经决策系数法和预测误差法验证,此模型的拟合精度和预测精度较高。将冲压工艺参数组合设置为蜜源位置,将参数优化问题转化为蜂群搜索最优蜜源问题。为了提高优化质量,对观察蜂的选择概率进行了改进,改善了算法的优化性能。经验证,改进蜂群算法的优化能力优于传统蜂群算法。使用优化参数得到的冲压件厚度分布均匀。使用优化后参数进行实际生产验证,冲压件最大减薄率为18.1%,与现行工艺参数相比,减薄率得到了有效降低,说明了冲压工艺优化的有效性和可实现性。

In order to decrease the maximum thinning rate of stamping part for vehicle bottom plate,an optimization method for process parameters based on improved bee colony algorithm was proposed. Taking blank holder force,stamping speed and die clearance as optimizing parameters,based on optimal Latin hypercube sampling method,the experiment was designed,and the experimental simulation result was obtained by finite element software Dynaform. Then,the response surface model of quality parameters and process parameters was established and verified by the decision coefficient method and predicting error method,and the fitting accuracy and predicting accuracy of the model were very high. Furthermore,by setting stamping process parameters as nectar source location,the optimization problem of parameters was transferred to the problem of bee colony searching for optimal nectar source. In order to improve optimization quality,the selection probability of observer bee was improved,and the optimization performance of the algorithm was advanced. It is verified that the optimization ability of the improved bee colony algorithm is better than that of the traditional bee colony algorithm,and the thickness distribution of stamping parts produced by the optimized parameters is uniform,and the maximum thinning rate of stamping parts is 18. 1%,which is reduced effectively compared with the current process parameters. Thus,the validity and realizability of stamping process optimization were verified.