Accurately reconstructing consecutive missing measurement segments presents a major challenge in bridge structural health monitoring, due to the inherent complexity and non-stationary nature of the collected signals. To address this, a hybrid data recovery framework is developed by integrating Improved Complete Ensemble Empirical Mode Decomposition with Adaptive Noise, Permutation Entropy, Long Short-Term Memory networks, and the Informer model (ICEEMDAN-PE-LSTM-Informer). The process begins by employing ICEEMDAN-PE to decompose the raw monitoring data into multiscale intrinsic mode components with distinct temporal characteristics. Subsequently, a parallel LSTM-Informer architecture is designed to forecast each decomposed component independently. This dual-branch design combines the local temporal modeling capability of LSTM with the long-range dependency modeling capability of Informer, and the extracted features are fused through a fully connected layer before prediction. The final recovered data sequence is obtained by linearly aggregating the predicted results from all components. The proposed method is evaluated using complete bearing-displacement and strain records from an operational bridge SHM system. Because the original datasets contain no missing observations, consecutive missing intervals are artificially constructed in the test sequences by controlled masking, allowing the original measurements to serve as ground truth for quantitative evaluation. Experimental results demonstrate improved reconstruction accuracy for the two bridge-monitoring datasets considered in this study. Specifically, for the longest prediction horizon of 72 steps, corresponding to a 12 h consecutive missing interval, the RMSE values for recovering bearing displacement and strain data are decreased by 54.06% and 44.11%, respectively, compared to a standard LSTM model, and by 46.06% and 31.47% compared to the standalone Informer model. These findings indicate that the proposed ICEEMDAN-PE-LSTM-Informer framework provides improved reconstruction performance over the considered baseline models for the two structural-response datasets investigated in this study.

Accurate identification of damage in simply supported girder bridges and the timely implementation of protective measures are crucial to preventing structural failure. Rotation influence line-based methods offer a straightforward and cost-effective approach for bridge monitoring; however, their validation has primarily relied on numerical simulations, with a lack of rigorous theoretical explanation. To address this limitation, an analytical relationship between the rotation difference at cross-sections before and after damage and the moving load position is first derived using the principle of virtual work, thereby clarifying the theoretical mechanism underlying damage identification. On this theoretical basis, a novel generalized information entropy index of rotation difference is proposed by incorporating information entropy theory to quantify the local nonlinear response induced by damage. The proposed method is validated through numerical simulations conducted on a simply supported steel girder bridge model developed in ANSYS, as well as through comparisons with existing experimental datasets. The results demonstrate that the proposed index can accurately and stably identify both single and multiple damage locations in bridges, while requiring only two inclination sensors installed at the supports. Under varying damage locations and severity levels, the entropy response curves consistently exhibit distinct peaks corresponding to the actual damage positions, thereby confirming the physical consistency and practical applicability of the method. The strategy of combining a minimal sensor configuration with information entropy analysis significantly reduces system complexity and cost while maintaining identification accuracy, providing an efficient and economical solution for practical bridge health monitoring.