关键词：传感器；桥梁（结构）；复合材料
摘 要：Structural Health Monitoring has great potential to provide valuable information about the actual structural condition and can help optimize the management activities. However, few effective and robust monitoring methods exist which hinders a nationwide use of SHM in structural condition evaluations. The objective of this project was to research and develop methods for structural identification and damage detection based on strain monitoring using long-gauge fiber-optic sensors. In particular two universal parameters of beam-like structures were studied in detail: the neutral axis and deformed shape. Data from two structures were used for validation purposes: from on-site monitoring of the US202/NJ23 overpass and from lab testing of a scale-model of a similar structure. The conclusions are that while the neutral axis varies during dynamic events, it changes the location due to damage, and it can be located accurately using a probabilistic approach. Thus, it can be used as a damage sensitive feature. At least two sensors per cross-section are necessary for an accurate evaluation of the location of the neutral axis. The vertical displacement of beams can be calculated by double integration of the curvature. However, the double integration method affects the accuracy of the evaluation, and to achieve the most accurate result a linear combination of integration methods is recommended. At least three locations along each girder should be instrumented with two parallel sensors for accurate evaluation of the vertical displacement. The methodologies researched in this project are presented in this report and recommendations for the use of the methods provided.