极端荷载作用下桥梁塑性损伤性能识别方法研究

Performance based design methods intend to set the structure components into plastic status when subjected to extreme loads (earthquake, vessel collision, etc.). As the key issue of bridge structural health monitoring, there are quite a few frequency and time domain parametric identification approaches available in the literature. However, most of these techniques have limitations practically in the following formats: none of the available methods are stably applicable to nonlinear elements; the time period required for data analysis is usually longer than the time length of the row data, indicating the manner of not real time or online analysis; the available system identification methods are capable of dealing relatively simple structures, instead of the complex structures. Thus, in this project, we proposed a new data analysis method, referred to as the adaptive quadratic sum-square error approach, for the on-line identification of elastic-plastic earthquake resistant designed bridge structures. The finite element model and equation of motion of the bridge structures will be established and the recursive solutions of the proposed approach will be derived. The accuracy and effectiveness of the proposed approach will be verified through simulation studies and experiments. Further, we propose a method of analyzing only partial data and updating the state vector using the rest data. In this way realizes the real time online analysis. Finally, the research outcome is widely applicable to the assessment of complex structures, i.e., bridge structures, subjected to earthquake or other accidental loads.

基于性能设计方法有意地设计结构在极端荷载（地震，船撞等）作用下进入可控的塑性状态，通过塑性构件（如塑性铰）或耗能构件（如阻尼器等）的减震效能作用来保证结构的安全性。在极端荷载发生时，通过监测到的结构响应及时识别结构损伤以及所设计的塑性构件和耗能构件的特性，对把握结构的性能、确认其可靠性、决定灾后恢复与加固方案，具有重要的意义。目前频域和时域范围的结构参数识别方法更多着眼于弹性构件的损伤(如刚度损失)，而对认为设计的弹塑性构件的损伤识别方法研究不够。在此背景下，本项目研究基于自适应最小误差法的损伤识别理论；研究静力凝聚方法与该方法的结合以缩减复杂结构中不易测量的自由度；对弹塑性抗震设计桥梁在极端荷载作用下的塑性损伤性能进行参数识别；讨论不同特性弹塑性模型的识别精度和适用性；并通过模型试验，验证方法的可行性。研究成果可广泛应用于桥梁等复杂结构在地震等突发荷载作用后的损伤评估。

目前工程设计有意地设计结构在极端荷载（地震，船撞等）作用下进入可控的塑性状态，通过塑性构件（如塑性铰）或耗能构件（如阻尼器等）的减震效能作用来保证结构的安全性。项目研究基于自适应最小误差法的损伤识别理论；研究静力凝聚方法与该方法的结合以缩减复杂结构中不易测量的自由度；对弹塑性抗震设计桥梁在极端荷载作用下的塑性损伤性能进行参数识别；讨论不同特性弹塑性模型的识别精度和适用性；并通过模型试验，验证方法的可行性。目前频域和时域范围的结构参数识别方法更多着眼于弹性构件的损伤(如刚度损失)，而对认为设计的弹塑性构件的损伤识别方法研究不够。研究提出的基于自适应最小二乘法的识别方法理论上能够兼顾全局及局部优化条件，试验和工程上能识别结构及构件的性能参数变化，且适用于给定非线性本构模型。当极端荷载发生时，通过监测到的结构响应及时识别结构损伤以及所设计的塑性构件和耗能构件的特性，对把握结构的性能、确认其可靠性、决定灾后恢复与加固方案，故该方法的发展具有重要的理论和工程实践意义。研究成果可广泛应用于桥梁等复杂结构在地震等突发荷载作用后的损伤评估，尤其适用于快速评估。