Innovated bridge health diagnosis model using bridge critical frequency ratio

The current bridge routine detection method in Taiwan relies on DER&U visual inspection, emphasizing ease and time efficiency. However, its accuracy is contingent on inspectors' experience and fails to assess internal pillar damage from external attacks. The prevalent direct approach in Taiwan Highway Administration, while obtaining dynamic bridge properties, involves mounting vibration sensors directly on the bridge, incurring significant time and cost. This research introduces an indirect approach, enhancing portability and cost-effectiveness by installing sensors on vehicles rather than bridges. To establish a bridge health standard, the study develops bridge models and conducts pushover analyses to speculate on the variation of the bridge vibration frequency ratio ( \({R}_{ec}\) $R_{\mathit{ec}}$ and \({R}_{sc}\) $R_{\mathit{sc}}$ ). This ratio serves as a crucial reference for determining bridge safety. To mitigate time and cost constraints, the research employs a hybrid symbiotic organisms search-least squares support vector machine (SOS-LSSVM) for \({R}_{ec}\) $R_{\mathit{ec}}$ and \({R}_{sc}\) $R_{\mathit{sc}}$ . The results enable the determination of vibration frequency ratios for the bridge safety standard. The indirect approach proves valuable for the Taiwan Highway Administration, allowing bridge frequency measurement under normal circumstances and post-disaster, facilitating timely decisions on bridge openings and providing a reference for regular maintenance based on bridge health assessments.