Despite their profound cultural and artistic significance, immovable stone cultural relics face severe threats from both environmental factors and human activities. Beyond these gradual damages, sudden events like impact loading pose an even greater risk and often result in catastrophic failure. To address this, a high-fidelity discrete element model of a Buddha statue was established to evaluate its dynamic response under impact conditions. Through six falling heights, this study examined the damage patterns and failure modes under different impact intensities. The results reveal that tensile failure dominates the damage process, with the kinetic energy from the falling ceiling slab first converting into strain energy within the statue and then dissipating rapidly after the formation of through-going fractures. Two critical falling heights were identified, allowing for quickly assessing the risk level. By integrating precise geometric and material data, the proposed method is highly adaptable to diverse heritage preservation tasks.

This paper highlights the effect of non-plastic fines on sand matrix by performing a series of undrained triaxial tests. The silty sand specimens were reclaimed by moist tamping method and two distinct density indexes of sand matrix were selected: I_Dmat=0.00 (0.40) for medium-dense state. The addition of non-plastic fines to the host sand matrix makes the specimens more resistant to liquefaction by inducing dilatant behaviour and reducing the excess pore water pressure Δu. As for instability line in the very loose state (I_Dmat=0.00), a rotation is observed on the p^’-q plane with increasing non-plastic fines fraction. This trend is non-linear and becomes less remarkable after a certain threshold value in fines fractions. In the medium-dense state (I_Dmat=0.50), there is no significant effect of non-plastic silt particles on the phase transformation state. Regarding the ultimate steady state, the line is unified in the p’-q space, but a clockwise rotation is observed in the log(p’)-e space. Furthermore, the exponential form shows a good fitting performance for silty sand.

Bridge failures inevitably cause severe social consequences in terms of the economic losses or even human fatalities. Scour is one of the leading causes for bridge failures. Due to its complexity and multidisciplinary nature, the mechanism of scour has not been completely understood yet. Inspired by the rapid development of Machine Learning (ML) techniques, this paper aims to construct a novel data-driven extreme gradient boosting (XGBoost) algorithm based model to predict the scour risk around bridge piers. Data used in the present study is provided by the French National Railway Company (SNCF). The performance of XGBoost-based model is compared with three commonly used algorithms: support vector machine, random forest and multilayer perceptron. Results show that the XGBoost classifier achieves high accuracy (0.959/0.938), precision (0.970/0.961), recall (0.974/0.956) and low false positive rate (0.085/0.114) for training and test set respectively. Moreover, the classifier obtains an area under the ROC curve (AUC) score equal to 0.974 for test set (a perfect classifier has an AUC equal to 1). This paper presents a cutting-edge application of XGBoost algorithm in the maintenance of railway bridges. The proposed methodology may allow engineers to determine the scour risk of bridge piers in an accurate and rapid way.

This article highlights the development of a large shaking table test for sand liquefaction analysis. Two soil containers of different sizes were fabricated. The first one was small (0.87 m × 0.87 m × 1.20 m) in which the reconstitution and saturation methods could be easily tested. The dry tamping (DT) method was used to fabricate a model specimen. The subsequent field measurements suggested that the DT method provided a good distribution of sand grains in different cross sections. Before supplying the model specimen with water, carbon dioxide was flushed to replace air bubbles. This helped in obtaining a good degree of saturation, later verified by a digital moisture meter. For a given inlet water flux, the recorded pore water pressure displayed a quasi-linear trend, suggesting a good internal void system. This reconfirms the effectiveness of the DT method to yield homogeneous model specimens. The second soil container was huge (4 m × 4 m × 2 m) and used to explore liquefaction behavior in real engineering dimensions. Flexible foams were mounted on the side walls to mitigate the boundary effect. Although the boundary effect still manifested itself near the edges of the container during base shaking, half of the model specimen underwent a correct simple shear condition. For further analysis, vane shear tests were carried out before and after the liquefaction test. It was found that the intermediate layer, in general, suffered from the most severe liquefaction failure.