基于GA优化ANN方法的端面机器人模锻参数优化  

Optimization of die forging parameters of end face robot based on GA optimization ANN method

作　　者：冯磊[1] FENG Lei(Taizhou Higher Vocational and Technical School of Mechanical and Electrical Engineering,Department of Mechanical and Electrical Engineering,Taizhou 225300,China)

机构地区：[1]泰州机电高等职业技术学校机电系,江苏泰州225300

出　　处：《锻压装备与制造技术》2024年第3期99-101,共3页China Metalforming Equipment & Manufacturing Technology

摘　　要：端面机器人模锻控制系统是以气动柔顺执行器与力传感器来实现固定结构,通过激光扫描方法获得表面粗糙度,为机器人运动路线设计提供模锻轨迹。为了提高端面机器人模锻控制效率,建立了一种神经网络ANN和遗传算法GA相结合的最优机器人模锻工艺,之后利用对比实验完成GA优化ANN方法的可靠性验证。研究结果表明:经过GA优化ANN处理的方法获得了更高的加工效率,相对初始表面质量发生了一定程度的下降,通过调整权重系数获得更优加工性能。本研究有助于提高模锻效率,为后续的参数优化奠定一定的理论基础。The die forging control system of the end-facing robot uses pneumatic flexible actuator and force sensor to realize the fixed structure,and obtains the surface roughness by laser scanning method,which pro-vides the die forging trajectory for the robot's motion path design.In order to improve the control efficiency of die forging of end-facing robot,an optimal robot die forging process combining neural network ANN and genetic algorithm GA was established,and then the reliability of GA optimized ANN method was verified by comparative experiments.The results show that the GA-optimized ANN processing method can obtain higher machining efficiency and decrease the initial surface quality to some extent,and better machining performance can be obtained by adjusting the weight coefficient.This study is helpful to improve the die forging efficiency and lay a theoretical foundation for the subsequent parameter optimization.

关 键 词：端面机器人模锻 神经网络 遗传算法 参数优化 

分 类 号：TP24[自动化与计算机技术—检测技术与自动化装置]