Model Optimization of Artificial Neural Networks for Performance Predicting in Spot Welding of the Body Galvanized DP Steel Sheets

摘要

This paper focused on the performance predicting problems in the spot welding of the body galvanized DP steel sheets. Artificial neural networks (ANN) were used to describe the mapping relationship between welding parameters and welding quality. After analyzing the limitation existed in standard BP networks, the original model was optimized based on lots of experiments. Lots of experimental data about welding parameters and corresponding spot weld quality were provided to the ANN for study. The results showed that the improved BP model can predict the influence of welding currents on nugget diameters, weld indentation and the shear loads ratio of spot welds. The forecasting precision was so high that can satisfy the practical need of engineering and have some application value.