The Research on Parameter Optimization of Power Battery Pack Welding Based on Neural Network

摘要

In order to make the needs of welding process adapt to actual production, people expect to establish a relation model that used the least number of tests and experiment datum to guide the welding production. The parameters of resistance welding technology are optimized through artificial neural network combined with orthogonal test. Orthogonal test was arranged using the three factors and three levels for the design of experiment, and the range method was used to analyse the influence of related welding parameters on the quality of joints. Considering the difficulties of current welding scheme's selection and optimization, Some simple mathematical calculations and analyze on the effect of different welding process on the properties of solder joints is conducted by the neural network training. The results of experiment indicate that the method has the high predictive precision, and the max relative error does not surpass 5%. Establishment of the neural network model provides optimizing process parameters of welding and improves welding quality and efficiency.