The CO2 laser parameter optimization design and practical verification for a touch panel conductive film

This study investigated the touch panel conductive film processing quality by using a CO2 laser system. The processing material chooses ethylene terephthalate with a size of 10 cm×4 cm. Experimental results measuring the three characteristics included line width, heat-affected zone and bump height. It further analyzed the relationship between the control parameters of a CO2 laser cutting system and the quality characteristics. First, the orthogonal array in the Taguchi Method was applied to plan the experiment. The signal noise ratio (S/N) of the quality characteristics was calculated according to data obtained from the experiments. Second, data were pre-processed using the gray relation method and the processed data showed a relationship between the control parameters and various quality characteristics. Next, the fuzzy inference system was applied to solve the multiple quality characteristics to determine the optimal solution and the combinations of the optimal multiple laser control parameters. Moreover, the Back-Propagation Neural Networks (BPNN) with the Levenberg-Marquardt algorithm was applied to construct a prediction system and simulate experimental results. Finally, the optimal solution was experimentally verified. The results showed that the prediction error rate was within 5%, proving that the proposed prediction system can effectively predict the laser processing of transparent conductive film.