Spot welding is widely used for numerous industrial applications, especially in automobile, aerospace and electronic manufacturing applications. One of the most important issues is how to evaluate the nugget diameter of spot weld, upon which the quality of joint was found to be dependent, with non-destructive inspection method. This paper introduces the investigation on evaluating quality of spot weld using back-propagation network. The network is configured by learning the pattern sets which consist of electrical parameters detected during the welding process and the relevant nugget sizes (diameter and height) measured after welding. The electrical parameters composed of welding current and voltage between the electrodes are selected as the input of the network. The output of the network is the nugget sizes of the weld which are compared with the experimental data. Results showed that the neural network system is vaild for evaluating application in spot welding.
展开▼
