Dam displacement monitoring is crucial for assessing structural safety; however, conventional models often prioritize single-task prediction, leading to an inherent difficulty in balancing monitoring data quality with model performance. To bridge this gap, this study proposes a novel two-stage analytical framework that synergistically integrates an improved isolation forest (iForest) with a metaheuristic-optimized random forest (RF). The first stage focuses on data cleaning, where Kalman filtering is applied for denoising, and a newly developed Dynamic Threshold Isolation Forest (DTIF) algorithm is introduced to effectively isolate noise and outliers amidst complex environmental loads. In the second stage, the model’s predictive capability is enhanced by first employing the LASSO algorithm for feature importance analysis and optimal subset selection, followed by an Improved Reptile Search Algorithm (IRSA) for fine-tuning RF hyperparameters, thereby significantly boosting the model’s robustness. The IRSA incorporates several key improvements: Tent chaotic mapping during initialization to ensure population diversity, an adaptive parameter adjustment mechanism combined with a Lévy flight strategy in the encircling phase to dynamically balance global exploration and convergence, and the integration of elite opposition-based learning with Gaussian perturbation in the hunting phase to refine local exploitation. Validated against field data from a concrete hyperbolic arch dam, the proposed DTIF algorithm demonstrates superior anomaly detection accuracy across nine distinct outlier distribution scenarios. Moreover, for long-term displacement prediction tasks, the IRSA-RF model substantially outperforms traditional benchmark models in both predictive accuracy and generalization capability, providing a reliable early risk warning and decision-support tool for engineering practice.

Pulse-like ground motions can cause severe damage to buried cast iron (CI) pipelines, which necessitates the selection of optimal seismic intensity measures (IMs) to estimate pipeline repair rates. Such a selection is essential for mitigating uncertainty in the seismic risk assessment of buried CI pipelines. For the first time, this study systematically screens the optimal scalar and vector IMs for buried cast iron pipelines with lead-caulked joints under pulse-like ground motions by a symmetrical evaluation based on the criteria of efficiency, sufficiency, and proficiency, providing a new method for reducing uncertainty in pipeline seismic risk assessment. We initiate the study by selecting 124 pulse-like ground motions from the NGA-West2 database and identifying 19 scalar and 171 vector IMs as potential candidates. A two-dimensional soil–pipe model is introduced, incorporating variability in the sealing capacity of lead-caulked joints along the axial direction. CI pipeline repair rates are calculated across various scaling factors and apparent wave velocities, yielding 1116 datasets pertinent to CI pipeline damage. The repair rate is adopted as the engineering demand parameter (EDP) to evaluate the efficiency, sufficiency, and proficiency of candidate IMs. Through comprehensive analysis, peak ground velocity (PGV) and the combination of PGV and the time interval between 5% and 75% of normalized Arias intensity ([PGV, Ds_5–75]) are determined as the optimal scalar- and vector-IMs, respectively, for assessing the repair rate of buried CI pipelines under pulse-like ground motions.