基于贝叶斯更新的深水桥墩波浪动力响应概率模型

A BAYESIAN UPDATING BASED PROBABILISTIC MODEL FOR THE DYNAMIC RESPONSE OF DEEPWATER BRIDGE PIERS UNDER WAVE LOADING

海洋环境中的深水桥墩长期受到不规则随机波浪动力作用。传统设计规范中采用的特征波浪计算方法,虽然简单高效,但是无法充分考虑波浪随机性和动力效应对结构的影响。该文以深水桥墩在波浪作用下的动力响应与静力响应之间的概率关系为研究对象,应用贝叶斯更新理论建立简单有效的深水桥墩波浪动力响应概率模型。通过对比深水桥墩动力时程分析结果,验证模型的有效性,并对样本数量和模型形式进行讨论。研究表明:基于贝叶斯更新理论的深水桥墩波浪动力响应概率模型可以快速准确地计算波浪作用下深水桥墩的动力响应,能够有效修正传统静力方法存在的偏差,并可在样本数据非常有限的情况下对深水桥墩的动力响应进行较好估计。

Deepwater bridge piers suffer from long-term irregular and dynamic wave loading in ocean environment.Although the static wave calculation approach based on characteristic waves used in the traditional design practice is simple and efficient,the effect of randomness and dynamics on the structures cannot be considered.This paper focuses on the probabilistic relationship between dynamic response and static response of a deepwater pier under wave loading.A probabilistic model based on Bayesian updating was proposed.The accuracy of the probabilistic model was verified through the comparison with the dynamic results of time-domain analyses.The selection of sample size and the model formulations were then discussed.This research showed that the proposed probabilistic model could calculate dynamic response of the deepwater pier under wave loading quickly and efficiently and corrected the inherent bias of the conventional static method.The proposed model is also able to estimate the dynamic response of the deepwater pier even if the sample size is very small.