New Representative Temperature for Performance Alarming of Bridge Expansion Joints through Temperature-Displacement Relationship

It is important to understand the operational performance of bridge expansion joints in real time. This paper proposes a novel performance-alarming approach for bridge expansion joints using continuous temperature and displacement-monitoring data. It is first shown that the representative temperature used to establish the temperature-displacement relationship (TDR) model was actually a linear combination of temperature measurements collected from different monitoring locations. Then, a new type of representative temperature is proposed and referred to as the canonically correlated temperature, in which the combination coefficients were optimally determined to maximize the correlation between the new representative temperature and expansion joint displacement. A more accurate and reliable baseline TDR model was subsequently established through the canonically correlated temperature. After that, a performance-alarming approach for bridge expansion joints was formulated based on construction of a mean value control chart for the estimation error of the baseline TDR model. In addition, a new way to determine the control limits is presented to solve the problem related to the nonnormal distribution of estimation errors. Finally, an engineering application to a cable-stayed bridge was carried out. The results demonstrate that the proposed approach is superior to traditional ones in terms of modeling and prediction capabilities and is excellent for performance alarming of bridge expansion joints.