论文部分内容阅读
采用旋转中心复合设计进行了激光透射焊接热塑性塑料聚碳酸酯的实验设计,分别建立了响应曲面法和基于遗传算法的人工神经元网络法的数学模型,运用这两种数学模型建立激光透射焊接工艺参数(激光功率、扫描速度、夹紧压力、扫描次数)和连接强度的关系模型,然后使用这两种模型分别预测了连接强度,并优化了焊接工艺参数。系统地比较了两种模型的建模能力、泛化能力和优化能力。实验结果表明,两种优化方法的试验结果较接近,但是基于遗传算法的人工神经元网络的建模、泛化和预测的准确性比响应曲面的要好,因此遗传算法-神经元网络是优化焊接质量的一种更有效的方法。
The experimental design of laser-transmissive welding thermoplastic polycarbonate was carried out by rotating center composite design. The mathematical models of response surface method and artificial neural network based on genetic algorithm were established respectively. By using these two mathematical models, the laser transmission welding process Parameters (laser power, scanning speed, clamping pressure, scanning times) and connection strength, and then use these two models to predict the connection strength and optimize the welding process parameters. We systematically compared the modeling ability, generalization ability and optimization ability of the two models. The experimental results show that the two optimization methods are close to the experimental results, but the artificial neural network based on genetic algorithm modeling, generalization and prediction accuracy is better than the response surface, so the genetic algorithm - neural network is optimized welding A more efficient way of quality.