论文部分内容阅读
三维编织复合材料由于其材料结构及编织工艺的复杂性和众多工艺参数的影响，目前尚未建立成熟的力学模型。本文采用人工神经网络ＢＰ（ｂａｃｋｐｒｏｐａｇａｔｉｏｎ）算法，将编织工艺参数作为人工神经网络的输入，将弹性模量及强度性能作为输出，建立了编织工艺参数与力学性能的人工神经网络关系模型，并讨论了ＢＰ算法及网络构造。这种人工神经网络关系模型对于三维编织复合材料的实验、生产和应用，工艺参数的选取以及理论模型的研究都有重要的参考价值。本文最后给出了三维编织复合材料拉伸和压缩性能的实验结果与人工神经网络预测结果，通过对比显示，其模拟效果令人满意。
Due to the complexity of the material structure and the weaving process and the influence of many process parameters, the three-dimensional braided composites have not yet established a mature mechanical model. In this paper, artificial neural network BP (backpropagation) algorithm is used to make the weaving process parameters as the input of artificial neural network. The elastic modulus and strength performance are taken as output. The artificial neural network model of weaving process parameters and mechanical properties is established. BP algorithm and network structure. This artificial neural network relationship model has important reference value for the experiment, production and application of three-dimensional braided composites, selection of process parameters and theoretical model. Finally, the experimental results and artificial neural network prediction results of tensile and compressive properties of 3D braided composites are given. The simulation results are satisfactory.