Abstract
Abstract
Different types of sensors are installed in many structural health monitoring systems. Correlation between heterogeneous response quantities is often small, which makes it difficult to utilize all sensor data for damage detection. A possible solution to handle diverse response data in the time domain is to use autocovariance function (ACF) estimates instead of direct sensor data. If the excitation is stationary random, different response quantities yield functionally similar ACFs. Due to phase differences, applying only spatial correlation between different ACFs is insufficient, whereas spatiotemporal correlation is justified. Therefore, ACFs at different lags are used to estimate a spatiotemporal correlation model of the undamaged structure under different environmental or operational conditions. If the correlation model is violated, it is an indication of damage. A monitoring campaign was carried out using a finite element model of a bridge deck to validate the proposed method. Neither the excitation nor the environmental variables were measured. Noisy response was measured with displacement, velocity, acceleration, and strain sensors in different locations of the deck. Using diverse types of sensors resulted in almost similar damage detection performance than if only a single type of sensor was used.