【作者】 陈鑫； 邹波； 卢子兴；

【Author】 CHEN Xin, ZOU Bo, LU Zixing ( School of Aeronautic Science and Technology, Beijing University of Aeronautics and Astronautics, Beijing 100083, China)

【机构】 北京航空航天大学航空科学与工程学院；

【摘要】 本文采用ANN方法对微孔泡沫塑料拉伸本构关系进行了模拟和预测。首先,选取应变和应力值作为输入和输出量,并利用反向传播BP算法建立了全量型和增量型神经网络本构模型;然后,对微孔聚碳酸酯(PC)泡沫塑料的拉伸应力-应变曲线进行了模拟和预测。数值结果表明,BP神经网络方法可以较好地拟合聚碳酸酯微孔泡沫塑料的拉伸应力-应变曲线,并具有很强的泛化能力,可用于对微孔泡沫塑料拉伸本构关系的模拟和预测。更多还原

【Abstract】 The artificial neural networks(ANN) were used to simulate and predict the tensile constitutive models of microcellular polycarbonate (PC). Firstly strains and stresses were chosen as inputs and outputs, secondly incremental type and full type neural network constitutive models were derived according to back propagation (BP) algorithm. Thirdly the tensile stress-strain curves of microcellular PC foams were simulated and predicted. The predicted results show the tensile stress-strain curves of microcellular PC foams can be accurately fitted by BP neural network algorithm, and the predicted results are in good agreement with that of the experiments.更多还原