论文部分内容阅读
通过对各种疲劳门槛值的理论模型的分析 ,认为结构钢的拉伸性能指标和循环加载条件是裂纹体疲劳门槛值的决定因素 .建立了用屈服强度、抗拉强度、断裂延性和循环加载应力比预测疲劳门槛值的人工神经网络模型 ,并用 1 0种结构钢的 60个样本对该模型进行了训练 .结果表明 ,人工神经网络模型可以很好地描述疲劳门槛值与结构钢拉伸性能指标及应力比之间复杂的定量关系 .应用所训练的人工神经网络模型预测了部分结构钢的疲劳门槛值 ,预测的结果与实测值符合良好
Through the analysis of the theoretical models of various fatigue thresholds, it is considered that the tensile properties and the cyclic loading conditions of structural steel are the determinants of the fatigue threshold of cracked bodies, and the yield strength, tensile strength, fracture ductility and cyclic loading Stress ratio than the fatigue threshold value of the artificial neural network model, and the use of 10 kinds of structural steel 60 samples of the model was trained.The results show that the artificial neural network model can describe the fatigue threshold and structural steel tensile properties Index and stress ratio between the complex quantitative relationship between the application of trained artificial neural network model predicts the fatigue threshold of some structural steel, the predicted results in line with the measured values in good