基于人工神经网络的侵彻毁伤模式识别

Penetration Pattern Recognition Based on Artificial Neural Network

为了研究多因素作用下破片侵彻靶板的具体模式,利用人工神经网络方法,对输入参数进行识别,得到了相应的靶板毁伤模式。基于正交试验设计原则,利用ANSYS/LS-DYNA仿真出60组破片侵彻靶板的模式数据,作为神经网络训练的输入数据,对神经网络进行训练,另外选取3组数据作为验证数据对神经网络的训练效果进行验证。结果表明:该训练模式下,人工神经网络能够有效地识别多因素作用下破片对靶板的具体毁伤模式。

In order to study the specific pattern of fragmentation penetrating the target plate under the influence of multi factors,this paper uses the artificial neural network method to identify the input parameters and obtains the corresponding target damage model.Based on the principle of orthogonal experimental design,this paper uses ANSYS/LS-DYNA to simulate the model data of 60 sets of fragments penetrating into the target plate,and trains the neural network as the input data of neural network training.In addition,three sets of data are selected as the validation data to verify the training effect of the neural network.The results show that the artificial neural network can effectively identify the specific damage patterns of fragments to target plates under multi factors.