摘要
针对目前国内挤压模具寿命过低的情况,将有限元分析、神经网络和遗传算法结合起来应用到挤压模具型腔优化设计中。采用B样条函数插值描述凹模型腔轮廓形状,用有限元数值模拟获得型腔表面节点的应力场、速度场和温度场,基于修正Archard磨损模型计算型腔磨损深度,以此作为样本训练BP神经网络,建立模具型腔控制点与磨损深度之间的映射关系,再结合遗传算法以等磨损为目标,优化模具型腔轮廓形状。优化结果与序列二次规划法一致,可以降低模具磨损,提高模具寿命,结果与实际情况吻合,表明了这种设计方法是可行的。
Finite-element method, BP Neural Network and genetic algorithms were combined together to optimize extrusion die profile aimed at the case that extrusion die life is shorter in our country. A method of B-spline function interpolation was used to describe extrusion die profile. The temperature, pressure and velocity field of nodes that lie extrusion die profile were gained by FEM simulation. Wearing depth of extrusion die profile was calculated by modified Archard theory. The results were used to train BP neural network, so that nonlinear mapping relation between reference point of die profile and wearing depth was obtained. In order to carry out uniform wearing depth, genetic algorithm was applied to optimize extrusion die profile. The optimal result accords with the result of sequential quadratic programming method, which reduces wearing depth of extrusion die and improves die life, and the optimal result is in agreement with practical conditions, which shows that the design method is feasible.
出处
《润滑与密封》
CAS
CSCD
北大核心
2007年第1期56-59,共4页
Lubrication Engineering
基金
国家自然科学基金资助项目(50575097)
江苏大学高级人才基金项目(04JDG037)
关键词
挤压
模具型腔
磨损深度
优化设计
extrusion
die profile
wearing depth
optimum design
