Measures for decreasing errors of ANN models of spot welding quality monitor

The relations between monitoring information and spot welding quality faetors are substantially non-linear, and it will make model error increasing to describle such relation by linear model. To describle the complex relations more reasonably, the artificial neural network (ANN) theory is applied to spot welding process modeling. During process modelling, it is appeared that "false saturation" phenomenon is a main obstacle to decreasing errors of models. By analyzing this phenomenon, many valid measures to decrease the likelihood of "false saturation" are taken. The test results show that the measures to decrease errors of models are feasible and effective.