Today, bridge owners must consider increasing traffic demands (in both volume and weight) and also face concerns related to sustainability, resilience and liveability which were virtually unknown in the 1950s. Furthermore, legislators demand data-driven asset management decisions based on objective, quantitative and reliable bridge condition and performance evaluation. To explore the current state-of-the-art in objective performance and condition evaluation of constructed systems by leveraging technology, a 30-year old freeway bridge in New Jersey, exhibiting multiple complex performance deficiencies, was transformed into a field laboratory. To identify the root causes of performance concerns, Visual Inspection, Operational Monitoring, Forced Excitation Testing, Controlled Load Testing, Non-destructive Probes, Long-term Monitoring, Finite Element Modelling and Parameter Identification were conducted within a Structural Identification framework. The results showed that root causes of some performance deficiencies of the test bridge were identified definitively only through the application of field measurements and analyses integrated by following a scientific approach—i.e. Structural Identification. Controlled Proof-Load Testing was especially useful in demonstrating the location and impacts of damage and the remaining capacity although such an approach can only be considered for the most critical cases due to its high cost and disruption to operations. Operational monitoring was shown as a sufficient and much cheaper alternative for structural identification permitting the development of a 3D digital twin of the bridge, which proved critical in identifying the root causes of its deficiencies and formulating meaningful interventions. Without an a-priori model used for designing the experiments as well as a model (i.e. a digital twin) calibrated by parameter identification and used for simulations, it was not possible to offer options for corrective measures confidently. The study demonstrated the challenges in relying only on visual inspection when a multitude of interdependent mechanisms lead to damage and deterioration, and the information value of different experimental methods such as vibration testing, proof load testing, wide-area NDE scans and multi-year SHM in being able to understand the root causes of various damages.

中文翻译：

---

公路桥梁状况与性能评估技术

如今，桥梁所有者必须考虑不断增长的交通需求（无论是体积还是重量），还面临着与可持续性，韧性和宜居性有关的担忧，而这在1950年代几乎是未知的。此外，立法者要求基于客观，定量和可靠的过渡条件和绩效评估的数据驱动型资产管理决策。为了利用技术探索结构化系统的客观性能和状态评估的最新技术，新泽西州一座具有30年历史的高速公路桥梁表现出多种复杂的性能缺陷，现已改建为现场实验室。为了确定性能问题的根本原因，包括外观检查，运行监控，强制励磁测试，受控负载测试，无损探头，长期监控，在结构识别框架内进行了有限元建模和参数识别。结果表明，只有通过应用现场测量和分析方法并通过遵循科学方法（即结构识别）综合起来，才能确定出测试桥某些性能缺陷的根本原因。受控的载荷试验在显示损坏的位置和影响以及剩余容量方面特别有用，尽管这种方法由于成本高昂且操作中断而只能在最关键的情况下考虑使用。事实表明，操作监控是结构识别的一种足够便宜的替代方案，可以开发桥梁的3D数字孪生模型，事实证明，这对于确定缺陷的根本原因和制定有意义的干预措施至关重要。如果没有用于设计实验的先验模型以及通过参数识别校准并用于仿真的模型（即数字孪生模型），就不可能自信地提供纠正措施的选项。这项研究表明，当多个相互依赖的机制导致损坏和恶化时，仅依靠目视检查是一项挑战，并且还提供了不同实验方法（例如振动测试，证明载荷测试，广域NDE扫描和多年期SHM）的信息价值能够理解各种损害的根本原因。如果没有用于设计实验的先验模型以及通过参数识别校准并用于仿真的模型（即数字孪生模型），就不可能自信地提供纠正措施的选择。这项研究表明，当多个相互依赖的机制导致损坏和恶化时，仅依靠目视检查是一项挑战，并且还提供了不同实验方法（例如振动测试，证明载荷测试，广域NDE扫描和多年期SHM）的信息价值能够理解各种损害的根本原因。如果没有用于设计实验的先验模型以及通过参数识别校准并用于仿真的模型（即数字孪生模型），就不可能自信地提供纠正措施的选择。这项研究表明，当多个相互依赖的机制导致损坏和恶化时，仅依靠目视检查是一项挑战，并且还提供了不同实验方法（例如振动测试，证明载荷测试，广域NDE扫描和多年期SHM）的信息价值能够理解各种损害的根本原因。