The growing challenges posed by the deteriorating state of global infrastructure highlight the urgent need for advanced damage detection methodologies. Structural Health Monitoring (SHM) systems play a critical role in this effort by facilitating real-time assessments of bridges through the collection of vast and diverse data. Proposing an innovative data driven approach for the anomaly detection on structural behavior, this study integrates a time series comparison metric and a temperature compensation strategy. The approach evidence feasibility on damage detection of a cable-stayed bridge.