| An approach of artificial neural network to robotic welding process modelling |
| |
| 引用本文： | Li Yan (Harbin Research Institute of WeldingJohn Norrish and T.E.B.Ogunbiyi Cranfield Institute of Technology,U.K.). An approach of artificial neural network to robotic welding process modelling[J]. 中国焊接, 1995, 0(1) |
| |
| 作者姓名： | Li Yan (Harbin Research Institute of WeldingJohn Norrish and T.E.B.Ogunbiyi Cranfield Institute of Technology U.K.) |
| |
| 作者单位： | Harbin Research Institute of WeldingJohn Norrish and T.E.B.Ogunbiyi Cranfield Institute of Technology，U.K. |
| |
| 摘 要： | Ａｎａｐｐｒｏａｃｈｏｆａｒｔｉｆｉｃｉａｌｎｅｕｒａｌｎｅｔｗｏｒｋｔｏｒｏｂｏｔｉｃｗｅｌｄｉｎｇｐｒｏｃｅｓｓｍｏｄｅｌｌｉｎｇ￥ＬｉＹａｎ（ＨａｒｂｉｎＲｅｓｅａｒｃｈＩｎｓｔｉｔｕｔｅｏｆＷｅｌｄｉｎｇＪｏｈｎＮｏｒｒｉｓｈａｎｄＴ．Ｅ．Ｂ...
|
An approach of artificial neural network to robotic welding process modelling |
| |
| Li Yan. An approach of artificial neural network to robotic welding process modelling[J]. China Welding, 1995, 0(1) |
| |
| Authors: | Li Yan |
| |
| Abstract: | Artificial neural networks(ANNs)have been investigated for application to robotic welding process.Two types of the ANN models are described.The first is a static modeling approach for the pre-setting of robotic welding parameters, and the other is a dynamic modelling for real time feedback control of robotic welding.These models map the relationship between the weld bead geometry and welding process parameters.Some basic concepts relating to neural networks are discussed. The performance of neural networks for modelling is discussed and evaluated by using actual robotic welding data.It is concluded that neural network is capable of modeling readily and quickly a multivariable welding process and the accuracy of neural networks modelling is comparable with the accuracy achieved by the statistical scheme. The choice between ANN and statistical models will depend on the application and control strategy used. |
| |
| Keywords: | artificial neural network weld modeling robotic welding |
| 本文献已被 CNKI 等数据库收录！ |
