基于神经网络的铝型材挤压过程能耗工艺参数优化研究  

Optimization of Extrusion Energy Consumption Parameters of Aluminum Profiles Based on Neural Network

作　　者：张明杰 杨柳 肖云 ZHANG Ming-jie;YANG Liu;XIAO Yun(College of Mechanical and Electronic Engineering,Guangdong University of Technology,Guangzhou 510006,China)

机构地区：[1]广东工业大学机电工程学院,广东广州510006

出　　处：《装备制造技术》2018年第6期258-259,共2页Equipment Manufacturing Technology

摘　　要：随着铝型材需求旺盛,能源消耗同步增长,高效低能耗挤压工艺研究是铝型材生产节能减排的重要方向。本文通过Hyper Xtrude分析软件对铝型材挤压过程进行有限元仿真,建立挤压过程能耗与工艺参数的神经网络模型,结合遗传算法优化工艺参数,找到最优参数组合,对降低挤压生产能耗及工艺参数选择有一定的指导意义。With the high demand of aluminum profile,the energy consumption is increasing synchronously,andthe research of high efficiency and low energy consumption extrusion technology is an important direction of energysaving and emission reduction of aluminum profile production. This paper through the analysis software of HyperXtrude finite element simulation of aluminum extrusion process,a neural network model of energy consumptionand extrusion process parameters,combined with the optimization of process parameters of genetic algorithm tofind the optimal combination of parameters,to reduce the energy consumption and the extrusion process parameters have certain guiding significance.

关 键 词：铝型材 能耗 神经网络 

分 类 号：TP319[自动化与计算机技术—计算机软件与理论]