论文部分内容阅读
提出改进响应面法研究在役PC桥梁承载力可靠性问题。利用神经网络模型拟合功能函数,采用JC法计算出桥梁结构构件的可靠指标,并进行了可靠性分析。工程实例分析表明:该方法建立的神经网络模型能够显示化表示复杂的桥梁承载力功能函数;随着使用年限的增长,桥梁上部结构的可靠指标计算值逐渐减小,说明上部结构抗力出现衰减;应用改进响应面法对在役PC桥梁承载力的可靠性计算是可行的。
An improved response surface methodology is presented to study the reliability of bearing capacity of PC bridge in service. The neural network model is used to fit the functional function, and the JC method is used to calculate the reliable index of the bridge structural components, and the reliability analysis is carried out. The engineering example shows that the neural network model established by this method can display the function function of complex bridge bearing capacity. With the increase of service life, the calculated value of the reliable index of the bridge superstructure decreases gradually, which shows that the resistance of the superstructure declines. It is feasible to use the improved response surface method to calculate the reliability of bearing capacity of in-service PC bridge.