Based on our prior study on uniaxial tearing tests, the influences of slit parameters, off-axis angles, and loading speeds on the tearing behaviors and strengths were discussed further. Results show that the tearing behaviors and strengths of the studied material are affected by above factors significantly, and typical tearing stress-displacement curves could be defined as four characteristic regions: a co-deformation region, a yarn extension region (or shear deformation region), a plateau region, and a post peak region. There is a W-shaped relationship between tearing strength and off-axis angle, with a small peak at 45°, and a new tearing strength criterion consisting of two parts, i.e., an inverted V-shaped part and a U-shaped part, was proposed for this laminated fabric. Comparisons between the experimental and the calculated results for the laminated fabric were performed, and the strong agreements demonstrate that the proposed criterion is feasible and accurate.

The glassy carbon electrode (GCE) was modified by electrochemically reduced graphene oxide (ERGO) and polypyrrole (PPy) prepared by simple cyclic voltammetry (CV) electropoly­merization. The PPy/ERGO modified electrode (PPy/ERGO/GCE) was used as a platform of electrochemical sensor to detect imidacloprid (IMI) insecticide. CV and differential pulse voltammetry (DPV) were chosen as the methods to investigate of the electrochemical behavior of IMI on PPy/ERGO/GCE surface. Scanning electron microscopy (SEM) and Raman spectra were utilized to describe the morphology and structure of the modified electrode. Experimental parameters were optimized, such as the number of polymerization cycles, scan rate and the pH value of electrolyte. Under the optimized conditions, when the concentration of IMI was in the range of 1-10 μM and 10-60 μM, the increase of reduction peak current was linear with the concentration of IMI, and the low detection limit was found to be 0.18 μM (S/N = 3). Results showed that PPy/ERGO/GCE demonstrated satisfactory reproducibility and stability, and has great potential in actual sample testing.

The construction of a reasonable and reliable deformation prediction model is of great practical significance for dam safety assessment and risk decision-making. Traditional dam deformation prediction models are susceptible to interference from redundant features, weak generalization ability, and a lack of model interpretation. Based on this, a deformation prediction model that considers the lag effect of environmental quantities is proposed. The model first constructs a new deformation lag influence factor based on the plain HST model through the lag quantization algorithm. Secondly, the attention and memory capacity of the model is improved by introducing a multi-head attention mechanism to the features of the long-time domain deformation influence factor, and finally, the extracted dynamic features are transferred to the ConvLSTM model for learning, training, and prediction. The results of the simulation tests based on the measured deformation data of an active dam show that the introduction of the deformation lag factor not only improves the interpretation of the prediction model for deformation but also makes the prediction of deformation more accurate, and it can improve the evaluation indexes such as RMSE by 50%, the nMAPE by 40%, and R² by 10% compared with the traditional prediction model. The combined prediction model is more capable of mining the hidden features of the data and has a deeper picture of the overall peak and local extremes of the deformation data, which provides a new way of thinking for the dam deformation prediction model.

In this study, a distributed optical fiber sensing technique based on Brillouin optical time‐domain analysis (BOTDA) is used to construct a complete bridge health monitoring system by continuously laying distributed sensing fiber lines in a steel box girder. The bridge scaffolding dismantling process is monitored to study the variation of the strain distribution. Additionally, a bridge finite element model is built to simulate the bridge scaffolding removed condition, and the strain distribution of the long‐span steel box girder viaduct after scaffolding dismantling is compared with the measured values. This study provides a reference for monitoring the scaffolding dismantling process based on BOTDA technology.

To improve the reliability and physical consistency of time-series prediction in safety-critical engineering systems, this study proposes a unified physics-constrained learning framework for sequential modeling. Unlike conventional data-driven approaches that rely solely on statistical fitting, the proposed method embeds domain-specific physical knowledge into neural networks through a structured and differentiable constraint integration strategy. Eight physically interpretable constraints are formulated to characterize key deformation mechanisms, including temporal evolution, environmental response, and intrinsic dynamic behaviors such as rate regulation and seepage–deformation interaction. These constraints are incorporated as normalized penalty terms and jointly optimized with data-driven objectives, enabling the model to learn nonlinear temporal dependencies while maintaining physically plausible predictions. To further investigate the effectiveness of constraint design, a cumulative experimental scheme is developed to analyze the marginal contributions and interaction effects among constraints. The results reveal a stepwise performance improvement with minor intermediate fluctuations, highlighting the nonlinear coupling and synergy between different physical mechanisms. The proposed framework is validated using long-term monitoring data from a real-world dam system and compared with classical statistical models and advanced deep learning methods. Experimental results demonstrate consistent improvements in prediction accuracy (R², RMSE, MAE, nMAPE) and trend consistency (Kendall’s τ), along with significant suppression of non-physical oscillations. Statistical significance tests further confirm the robustness of the performance gains. The framework is general and can be extended to other physics-constrained time-series prediction tasks in engineering systems. This study provides a practical and interpretable solution for integrating physical knowledge into data-driven modeling.