【摘要】 以易切削黄铜的加工表面粗糙度与各种加工参数的关系为对象,将L9( 34)型正交切削试验数据作为训练学习样本,同时以与正交试验参数有关的6个样本作为预测样本,用BP神经网络对其进行了预测。结果表明:经设计的BP神经网络训练1183次,其最大误差不超过5 % ;人工神经网络与正交试验相结合,能大大节省预测时间和费用,效果很好。更多还原

【Abstract】 The fundamental of the artificial neural network is introduced. The studied object was considered as the relations between machining parameters and the surface roughness of free-cutting brass and the trained samples considered as the data of the L 9(3 4) orthogonal test, the random six samples related to the orthogonal test were forecast in the BP artificial neural network. The results of tests show that the maximum error was 5% after 1183 trainings. Therefore, the practice was proved that the time and expense can greatly been saved and the effect of forecasting was very good if the artificial neutral network was in combination with the orthogonal test.更多还原