基于遗传算法和神经网络的冷挤压工艺参数模糊优化设计  

FUZZY OPTIMIZATION OF COLD EXTRUSION MOLD BASED ON GENETIC ALGORITHM AND NEURAL NETWORK

作　　者：石峰[1] 娄臻亮[1] 张永清[1] 陆金桂[2] 

机构地区：[1]上海交通大学模具CAD国家工程研究中心,上海200030 [2]南京化工大学

出　　处：《机械工程学报》2002年第8期45-49,共5页Journal of Mechanical Engineering

基　　金：教育部优秀青年教师基金;国家自然科学基金(60175019);上海市高等学校青年科学基金资助项目

摘　　要：将模糊优化思想引入冷挤压工艺参数优化设计,以冷挤压模具的四个关键工艺参数为设计变量,以最终挤压件的各处材料的损伤值为目标函数,建立了冷挤压工艺参数模糊优化模型。提出了利用神经网络进行材料损伤近似计算的策略,从而形成了基于遗传算法和神经网络的冷挤压工艺参数模糊优化方法。对冷挤压工艺参数模糊优化模型进行了求解,优化结果表明模糊优化思想能提高冷挤压工艺设计质量。The idea of fuzzy optimization is introduced in the optimization of cold extrusion mold process parameters. An approach of fuzzy optimization of cold extrusion mold process parameters based on genetic algorithm and multiplayer neural network is presented. Firstly, the fuzzy optimization model of cold extrusion mold is introduced. Secondly, the fuzzy optimization strategy using genetic algorithm and multiplayer neural network is discussed, including constructing an approximate model with artificial neural networks and solving the fuzzy optimization model using genetic algorithms. In constructing fitness function, the weight factors are determined by orthogonal numerical experiments. Finally, the optimization results demonstrate the fuzzy optimization method can improve the quality of the cold extrusion mold.

关 键 词：遗传算法 模糊优化 神经网络 冷挤压 工艺参数 正交试验 优化设计 

分 类 号：TG376.3[金属学及工艺—金属压力加工] TP18[自动化与计算机技术—控制理论与控制工程]