An important issue in structural engineering is improving performance and safety by retrofitting/strengthening or repairing of existing reinforced concrete structures. Attaching a steel bracing system is one of the common solutions for retrofitting or repairing of damaged structures. An efficient load transfer between existing structural members and an attached retrofit system is significant to such repairs. For steel bracing systems, the load transfer between the steel system and existing concrete structures are achieved through proper anchorage between them. A typical steel-to-concrete connection includes a steel attachment - consisting of a baseplate welded to the attached member - and the anchors that actually connect the steel to concrete as bolting system. In the present paper, the reliability method is used to assess the safety of the steel bolting system that is used to connect steel frame systems to existing reinforced concrete columns. Herein, the steel frame systems are used to retrofit/strengthen existing reinforced concrete corbel beams as a structural support system. The effect of applied loadings, bolts geometry, bolts spacing variation and concrete quality on the concrete breakout strength are examined for the current problem. Reliability is assessed by estimating the failure probability of the single bolt failure and system bolt failure. The results are used to determine the parameters affecting safety of the connection system and to improve the performance and the technique of strengthening/repairing of the steel supporting frame systems.

Severe ice accretion on transmission lines can disrupt electrical grids and compromise the stability of power systems. Consequently, precise prediction of ice coating on transmission lines is vital for guiding their operation and maintenance. Traditional single-model icing prediction methods often exhibit limited accuracy under varying environmental conditions and fail to yield highly accurate predictions. We propose a multi-scenario, two-stage adaptive ensemble strategy (MTAES) for ice coating prediction to address this issue. A combined clustering approach is employed to refine the division of ice weather scenarios, segmenting historical samples into multiple scenarios. Within each scenario, the bagging approach generates multiple training subsets, with the extreme learning machine (ELM) used to build diverse models. Subsequently, a two-stage adaptive weight allocation mechanism is introduced. This mechanism calculates the distance from the scenario cluster centers and the prediction error of similar samples in the validation set for each test sample. Weights are dynamically allocated based on these data, leading to the final output results through an adaptive ensemble from the base model repository. The experimental results show that the model is significantly better than traditional models in predicting ice thickness. Key indicators of RMSE, MAE, and ${\mathrm{R}}^2$ reach 0.675, 0.522, and 83.2%, respectively, verifying the effectiveness of multi-scene partitioning and adaptive weighting methods in improving the accuracy of ice cover prediction.

Structural health monitoring (SHM) under realistic operating conditions requires damage indicators that are physically interpretable and robust to uncertain excitation. This paper proposes a physics-informed indicator for structural damage detection based on local polynomial representations of the measured response. The approach approximates the structural motion within short time intervals using a series expansion, conceptually related to differential transform representations. The governing equations are recast in a term-wise form to derive closed-form relationships between the local coefficients and the system parameters. For linear systems, the ratio of the second-order to the zeroth-order coefficient gives an exact local estimator of stiffness in the undamped and unforced limit, and remains a consistent estimator under light damping. For nonlinear systems with cubic stiffness, higher-order coefficients encode nonlinear effects through explicit algebraic coupling, allowing the definition of nonlinearity-sensitive features. A sensitivity analysis shows that the proposed features depend monotonically on stiffness degradation and vary systematically with nonlinear parameters. Closed-form error bounds quantify the influence of damping and external forcing, showing that the proposed features stay accurate under light damping and bounded excitation. The method is validated through numerical studies on linear and nonlinear vibration systems. The numerical experiments introduce variations in stiffness and nonlinear parameters. The results show that the proposed features track the parameter changes accurately and separate linear from nonlinear dynamic effects.