当前位置: X-MOL 学术Struct. Infrastruct. Eng. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Determination of appropriate updating parameters for effective life-cycle management of deteriorating structures under uncertainty
Structure and Infrastructure Engineering ( IF 3.7 ) Pub Date : 2020-08-24 , DOI: 10.1080/15732479.2020.1809466
Baixue Ge 1 , Sunyong Kim 1
Affiliation  

Abstract

Effective life-cycle management of deteriorating structures should be based on an accurate and reliable damage propagation prediction model under uncertainty. Appropriate updating with inspected information can decrease errors between the detected and predicted damage and finally improve the accuracy and reliability of life-cycle management of a deteriorating structure. This paper presents an approach to determine the most appropriate probabilistic parameters to update the damage propagation prediction model. The presented approach includes the comparison-based method and parametric global sensitivity analysis (GSA). In the comparison-based method, the most appropriate updating parameters are determined based on the mean absolute error (MAE), Kullback–Leibler (KL) divergency, and Bhattacharyya distance by considering all the combinations of probabilistic parameters related to damage propagation prediction model. Furthermore, the GSA can be also applied to select the most appropriate parameters where sensitivity indices of the combination of the probabilistic parameters are compared. Finally, a comparison of assessment values for the most appropriate parameters selected from comparison-based method and GSA is presented. The approach presented in this paper can be applied for any damage propagation model. Superstructures of two existing bridges under corrosion and fatigue are applied as illustrative examples.



中文翻译:

确定适当的更新参数,以在不确定性条件下对老化结构进行有效的生命周期管理

摘要

老化结构的有效生命周期管理应基于不确定性下准确可靠的损伤传播预测模型。对检查信息进行适当的更新可以减少检测到的和预测的损坏之间的误差,最终提高老化结构生命周期管理的准确性和可靠性。本文提出了一种确定最合适的概率参数以更新损伤传播预测模型的方法。所提出的方法包括基于比较的方法和参数化全局敏感性分析 (GSA)。在基于比较的方法中,根据平均绝对误差 (MAE)、Kullback–Leibler (KL) 散度确定最合适的更新参数,和 Bhattacharyya 距离通过考虑与损伤传播预测模型相关的概率参数的所有组合。此外,还可以应用 GSA 来选择最合适的参数,其中比较概率参数组合的灵敏度指数。最后,介绍了从基于比较的方法和 GSA 中选择的最合适参数的评估值的比较。本文提出的方法可以应用于任何损伤传播模型。以腐蚀和疲劳状态下的两座现有桥梁的上部结构为例。GSA 也可用于选择最合适的参数,其中比较概率参数组合的灵敏度指数。最后,介绍了从基于比较的方法和 GSA 中选择的最合适参数的评估值的比较。本文提出的方法可以应用于任何损伤传播模型。以腐蚀和疲劳状态下的两座现有桥梁的上部结构为例。GSA 也可用于选择最合适的参数,其中比较概率参数组合的灵敏度指数。最后,介绍了从基于比较的方法和 GSA 中选择的最合适参数的评估值的比较。本文提出的方法可以应用于任何损伤传播模型。以腐蚀和疲劳状态下的两座现有桥梁的上部结构为例。

更新日期:2020-08-24
down
wechat
bug