Parameter Optimization of Laser Scribing Technics of 30Q130 Grain-Oriented Silicon Steel Based on Genetic Neural Network

摘要

A laser is often considered to scribe the grain-oriented silicon steel surfaces after cold-rolling and annealing to reduce the core loss. It is necessary to select the best scribing parameters to maximize the reduction in this process. This paper proposed an optimization method of genetic algorithm during laser scribing of 30Q130 steel, by developing an artificial neural network prediction model using a database form a designed orthogonal experiment. The objective was to determine the best combination values of three important scribing parameters, namely scribing velocity, pulse energy and scanning spacing, that can get the largest core loss reduction. An optimized combination of parameters was obtained by this method and then validated by an adding experiment. The result indicates that the optimization model is reliable.