Stability evaluation of geotechnical engineering slopes is of great significance for the risk control and safe operation of many engineering. Machine learning methods can effectively establish the potential relationship between geological features and slope behavior under complex environments, to accurately evaluate the stability of slope rock and soil. This work investigated the performance of eight commonly used machine learning models to predict slope safety factors. First, the prediction system of slope safety factors based on machine learning was established by combining historical data of slopes for parameter optimization and cross-validation. Then, four accuracy evaluation indexes, MSE, RMSE, MAE, and Pearson correlation, were objectively weighted, and objective weighting-TOPSIS models were constructed to comprehensively quantify the performance of each model. Finally, the best machine learning model was used in the slope stability analysis of the Sino–Russian natural gas control section. The research results show that there are obvious differences in the prediction accuracy of the slope safety factor among different models. The ANN model has the highest evaluation accuracy, and the ensemble learning method performs well in the data set. The machine learning model can better predict the safety factor of the slope under different working conditions. The discrepancies with the numerical simulation results are related to the limitations of data sets and the differences in analysis methods. The analysis method of this study not only provides a new research idea and solution for the construction and evaluation of the model predicting slope safety factors, but also applies to other geotechnical engineering instability problems.

Slope surface deformation monitoring plays an important role in landslide risk assessment and early warning. Currently, the mainstream GNSS, as a point-measurement technique, is expensive to deploy, resulting in information on only a few points of displacement being obtained on a target slope in practical applications. In contrast, optical images can contain more information on slope displacement at a much lower cost. Therefore, a low-cost, high-spatial-resolution and easy-to-implement landslide surface deformation monitoring system based on close-range photogrammetry is developed in this paper. The proposed system leverages multiple image processing methods and monocular visual localization, combined with machine learning, to ensure accurate monitoring under time series. The results of several laboratory landslide experiments show that the proposed system achieved millimeter-level monitoring accuracy in laboratory landslide experiments. Moreover, the proposed system could capture slow displacement precursors of 5 mm to 10 mm before significant landslide failure occurred, which provides favorable surface deformation evidence for landslide monitoring and early warning. In addition, the system was deployed on a natural slope in Lanzhou, yielding preliminary effective monitoring results. The laboratory experimental results demonstrated the system’s effectiveness and high accuracy in monitoring landslide surface deformation, particularly its significant application value in early warning. The field deployment results indicated that the system could also effectively provide data support in natural environments, offering practical evidence for landslide monitoring and warning.

The Loess Plateau area, with complex geomorphological features and geological structure, is highly prone to geologic disasters such as landslides and debris flow, which cause great losses. To investigate the initiation mechanism of landslide and debris flow disasters and their spreading patterns, historical satellite images in the Laolang gully were collected and digitized to generate three-dimensional topographic and geomorphological maps. Typical landslides were selected for landslide thickness measurement using a standard penetrometer and high-density electrical method. Numerical models were established to simulate the occurrence and development of landslides under different working conditions and to evaluate the spreading range based on the propagation algorithm and friction law. The results show that the 10 m resolution DEM data are well matched with the potential hazard events observed in the field site. The smaller the critical slope threshold, the greater the extent and distance of landslide spreading. The larger the angle of arrival, the greater the energy loss, and therefore the smaller the landslide movement distance. The results can provide scientific theoretical guidance for the prevention and control of rainfall-induced landslide and debris flow disasters in the Loess Plateau area.

Remote micro-deformation monitoring of large structures and geological hazards presents significant challenges for conventional phase-interferometric radar due to limitations in range, transmitter leakage, and multi-target identification. This study proposed a deformation monitoring radar utilizing a frequency-multiplication transponder. Frequency isolation effectively suppresses transmitter leakage, while active frequency-multiplication amplification enhances echo signals over long distances. Additionally, on–off keying (OOK) frequency division allows for the separation of multiple targets. A transponder interferometric model established a relationship between target displacement and echo phase. A prototype-measured link budget established the theoretical operational range limit. Each measurement point required a dedicated active transponder. Consequently, the system conducted sparse point-wise radial displacement monitoring rather than full-field imaging. Field experiments were performed to evaluate displacement measurement at 129 m, 345 m, and 570 m. A dual-target experiment was also conducted at 129 m. All test configurations successfully resolved 2.5 mm micro-displacements. The measurement error was approximately 0.1 mm under unobstructed conditions. The maximum absolute error remained below 0.8 mm in complex environments. Dual-target demodulation yielded a root-mean-square error of 0.14 mm. The maximum experimentally verified distance is 570 m. The 1 km design range represents a theoretical link-budget target. This target is constrained by the transponder saturation threshold and receiver sensitivity. Field validation at 1 km remains unachieved.