The Use of the Taguchi Method and a Neural-Genetic Approach to Optimize the Quality of a Pulsed Nd:YAG Laser Welding Process



In the production process of lithium-ion secondary batteries, the lap-joint quality of the safety vent and the cathode lead influences the product quality and production effciency. A pulsed Nd:YAG laser welding machine was employed herein. The welding parameters that influence the pulsed Nd:YAG laser welding quality was evaluated by measuring the tensile-shear strength. In this study, the Taguchi method was used to perform the initial optimization of the process parameters. A neural network (NN) with the Levenberg–Marquardt back-propagation algorithm was adopted to develop the nonlinear relationship between factors and the response. Then, a genetic algorithm based on a well-trained NN model was applied to determine the optimal factor settings. Experimental results illustrated the proposed approach.