The lack of reliable negative samples is an important factor limiting the quality of machine learning-based debris flow susceptibility mapping (DFSM). The purpose of this paper is to propose multiple negative-sample acquisition strategies for DFSM considering different sample representation forms. The sample representation forms mainly include a single grid, multi-grid, and watershed unit, and the negative-sample acquisition strategies are based on support vector machine (SVM), spy technique, and isolation forest (IF) methods, respectively. These three strategies can assign a value to all the samples based on different assumptions, and reliable, negative samples can be generated from samples with values below a predefined threshold. Combining different sample representation forms with negative sample acquisition strategies, nine datasets were then involved in random forest (RF) modeling. The receiver operating characteristic (ROC) curves and related statistical results were used to evaluate the models. The results show that the strategy based on the spy technique is suitable for multiple datasets, while the IF-based strategy is well-adapted to the watershed unit datasets. This study can provide more options for improving the quality of datasets in DFSM, which can further improve the performance of machine learning models.

Debris flow hazard mapping (DFHM) played an important role in reducing the threat of debris flows. Conventional DFHM usually requires numerical simulations to obtain debris flow intensity, which is usually quite time-consuming. This paper is to introduce a combined application framework of numerical simulation and machine learning to improve the efficiency of DFHM. The FLO-2D model was employed to simulate debris flows with different recurrence intervals of 20 years, 50 years and 100 years, respectively. Corresponding maximum accumulation depth and flow velocity were collected as dependent variables. Rainfall with corresponding recurrence intervals, altitude, slope, plane curvature, profile curvature, topographic humidity index, normalized difference vegetation index and Manning’s coefficients were collected as independent variables. Then the 20-year, 50-year and 100-year datasets were prepared by combining the independent and dependent variables. The 20-year and 50-year datasets were used for training a machine learning model called gradient boosted decision tree (GBDT). The 100-year dataset was used for validation. The results showed that the predicted results are quite close to the simulated results, which verified the validity and rationality of the proposed method. In addition, the training and prediction process of machine learning models is more than 10 times faster than the running process of numerical model. This study proposed the potential use of machine learning models as alternatives to hydraulic simulations, which could provide a more efficient solution for large-scale DFHM.

The steel box girder bridge is a structure composed of mutually vertical stiffening ribs (longitudinal ribs and transverse ribs) that carry the loads of vehicles. Since the external loads are usually complex and variable, the rational design of the bridge components is a topic that deserves more attention. The purpose of this study is to explore the optimal range of some of the component design parameters, expecting to reduce costs while ensuring the stress-carrying capacity. A finite element model (FEM) based on ABAQUS was built and the results were verified by laboratory experiments. The varied thicknesses of the bridge deck, diaphragm, and U-rib were explored based on the validated FEM. The simulation results fit well with the experimental results, which proved that the FEM was quite reliable. The stress analysis results demonstrated an optimal range of 18–20 mm for bridge deck thickness, 14–16 mm for diaphragm thickness, and 8–10 mm for U-rib thickness. The present study holds significant reference value for the design and optimization of multiple steel box girder bridge components, which could further provide a theoretical foundation for related research in this field.

Light steel keel fireproof exterior walls are one of the new composite walls that have been widely used in recent years. This paper analyzes the factors affecting the displacements of the connectors and wall panels and modifies the corresponding displacement model. ABAQUS was used to establish the simulation analysis of the L-shaped and special-shaped steel angle component, and the results were compared with the experimental model to verify the rationality. The effect of vertical keel spacing and layout, the number of special-shaped steel angles, the L-shaped steel angle thickness, and the number of self-tapping screws for L-shaped steel angles to steel beams were explored based on the validated finite element model. The calculation model was modified based on the results of the simulation. The results showed that the simulation results fit well with the experimental results, which verified the reliability of the finite element model. The number of connectors and the number of self-tapping screws are the main factors affecting the special-shaped and L-shaped steel angle components, respectively. By modifying the rotation constraint factor of the L-shaped steel angle and adding a discount factor for the wall panel, the accuracy of the associated displacement calculation model could be improved.

The presentation of a constitutive model could help researchers to predict the mechanical behavior of a material, which also contributes to the further generalization of the material. This paper is to explore the tensile constitutive model of engineered cementitious composites (ECCs) reinforced by high-strength steel wire mesh based on experiments and numerical simulations. DIANA was used to simulate the tensile process of the specimens, and experiments were carried out to validate the numerical model. The effect of the ECCs’ tensile strength, reinforcement ratio and specimen size were considered during the specimen design process. The results showed that most of the errors of the simulated values compared to the experimental results were within 5%, which proved that the numerical model was quite accurate. The proposed constitutive model revealed the different roles played by ECCs and high-strength steel wires at different stress stages, and the calculation results were in high agreement with the simulation results, indicating the effectiveness of the constitutive model. The study in this paper could provide an important reference for the popularization and application of ECCs reinforced by high-strength steel wire mesh.