Application of neural networks in welding parameter planning of robots

被引:0
|
作者
Peng, Pai [1 ]
Wu, Lin [1 ]
Tian, Jin-Song [1 ]
Wang, Xue-Feng [1 ]
Feng, Ying-Jun [1 ]
机构
[1] Harbin Inst. of Technol., Harbin 150001, China
来源
| 2001年 / Harbin Research Institute of Welding卷 / 22期
关键词
Convergence of numerical methods - Dynamic programming - Neural networks - Parameter estimation - Welding;
D O I
暂无
中图分类号
学科分类号
摘要
Off line programming has become an important branch in robotic programming because it has many advantages such as wide applicability and not occupying production time. Welding parameter planning is one of necessary components in robotic arc welding task-level off line system. In order to perform welding parameter planning, a new training algorithm of the feedforward neural network known as single parameter dynamic search algorithm is utilized. Single parameter dynamic search is the characteristic of this algorithm and therefore the calculation quantum of objective function is reduced greatly. The idea of functional link is used to pretreat the input parameters of neural networks. The calculating results show that the convergence effect of this algorithm is better than that of BP algorithm in the field of welding parameter planning.
引用
收藏
相关论文
共 50 条