The accurate and reliable classification of rock mass is the basis of a reasonable engineering design. In the Xishan mining region of Sanshandao Gold Mine, three conventional rock mass classification methods of Tunneling Quality Index (Q), Rock Mass Rating (RMR) and China National Standard-basic quality (BQ), were compared in the burial depth area above 780 m, and it was discovered that the classification results of different rock mass classification methods had a low coincidence rate in the deep area; Therefore, this paper adopted entropy weight method, TOPSIS method and grey correlation analysis method to calculate the entropy weight and relative closeness of different methods in different middle sections. The study’s findings revealed that in the deep area, the relative closeness between each classification mass was: RMR > Q > BQ; Based on the above results, the IRMR method with modified RMR was selected for comprehensive analysis, and the concept of importance degree of evaluation index was defined; it was found that the importance degree of evaluation index of in-situ stress loss was the highest, while the importance degree of joint direction was the lowest; The “ETG” rock mass classification method based on “site-specific” is established, which provides a reference for the establishment of deep rock mass classification method.

This paper proposes a new method for predicting slope safety factors that combines convolutional neural networks (CNNs), gated recurrent units (GRUs), and attention mechanisms. This method can better capture long-term dependencies, enhance the ability to model sequential data, and reduce the dependence on noisy data, thereby reducing the risk of overfitting. The goal is to improve the accuracy of slope safety factor prediction, detect potential slope stability issues in a timely manner, and take corresponding preventive and control measures to ensure the long-term stability and safety of infrastructure and promote sustainable development. The Pearson correlation coefficient is used to analyze the relationship between the target safety factor and the collected parameters. A one-dimensional CNN layer is used to extract high-dimensional features from the input data, and then a GRU layer is used to capture the correlation between parameters in the sequence. Finally, an attention mechanism is introduced to optimize the weights of the GRU output, enhance the influence of key information, and optimize the overall prediction model. The performance of the proposed model is evaluated using metrics such as the mean absolute error (MAE), mean absolute percentage error (MAPE), mean squared error (MSE), root-mean-square error (RMSE), and R². The results show that the CNN-GRU-SE model outperforms the GRU, CNN, and CNN-GRU models in terms of prediction accuracy for slope safety factors, with improvements of 4%, 2%, and 1%, respectively. Overall, the research in this paper makes valuable contributions to the field of slope safety factor prediction, and the proposed method also has the potential to be extended to other time-series prediction fields, providing support for a wide range of engineering applications and further promoting the realization of sustainable development.

Strengthening the base-angle of a roadway can have a beneficial impact on its overall stability, but the specific optimal parameter selection range is still unclear. Fast Lagrangian Analysis of Continua 3D (FLAC3D) software is used to carry out the stability analysis of 5 kinds of roadway models with different section shapes under the conditions of no support and different base-angle bolt support angles, and the simulation verification is carried out under the actual working conditions of the Sanshandao Gold Mine. The conclusion is as follows: without support, the self-stability of a tri-centric arch roadway is the best, and that of a rectangular roadway is the worst. When the base-angle bolt dip angle is between 15° and 45°, a better supporting effect can be obtained under the established 5 kinds of roadway sections. In the straight wall circular arch roadway of the Sanshandao Gold Mine, the roadway stability is the best when the angle of the base-angle bolt is 15°. However, changing the length of the base-angle bolt, even if the length of the base-angle bolt is increased to a certain extent, will decrease the overall supporting effect of the supporting structure. This paper can provide a reference for the selection of tunnel bottom corner bolt support parameters.

The appropriate excavation sequence can improve the overall stability of a foundation pit. In this study, eight schemes were created using FLAC3D to examine the impact of the excavation sequence on a foundation pit by analyzing a deep foundation pit in Nanjing, which had an irregular large rectangle shape. The results show that different excavation sequence schemes and different phases of the foundation pit can change the displacement values and the horizontal displacement type. The min–max normalization method was used to score the schemes in terms of six parameters and confirm the best excavation sequence scheme. In addition, the irregular shape of the foundation pit also leads to local differences in the stability of a foundation pit; the wide end is only slightly longer than the narrow end, but its displacement is significantly higher than that of the narrow end, so attention should be paid to reinforcing the weak parts when carrying out the support. This study can inform the selection of the excavation sequence for actual construction processes.

Appropriate simulation set parameters are the precondition to obtain accurate results; while the simulation results are affected by multiple factors, it is thus crucial to investigate the sensibility of different factors. This paper first analyses the application situation of numerical simulation software in the field of geotechnical engineering and finds that Fast Lagrangian analysis of continua in three dimensions (FLAC3D) has been widely used on roadways or tunnels. Then, taking the roadway excavation process as the engineering background, FLAC3D was used to create 171 schemes of different simulation parameters and analyze the influence of different factors on the simulation results. The findings show that there is a considerable difference in the degree of effect of different parameters on the simulation results. Most of the factors have a remarkable effect on the numerical simulation results (displacement and stress), and only some factors (parameter uniformity and density) have almost no effect on the results. Meanwhile, the trend of displacement and stress is opposite in most cases. In addition, some neglected factors can also have a considerable effect on the simulation results, such as the zone amount; therefore, it is necessary to avoid the variation of nonstudy factors as possible when carrying out the numerical simulation. This study may significantly assist concerned engineers and technicians in developing a more organized and thorough grasp of the impacts of various parameters on simulation outcomes.

The time and space of the support structure applying is related to the overall stability of the roadway after excavation directly. Designed twenty-four groups of roadway support schemes with time and space dimensions, and studied the stability characteristics of roadway in different schemes by using Fast Lagrangian Analysis of Continua in Three Dimensions. The main conclusions are as follows: the influence of energy dissipation (time) and support position (space) on the stability of the roadway is not a linear relationship, and supporting at the appropriate opportunity can be beneficial. Established the "displacement-dissipation energy" curve, founding that there is an obvious "jump" phenomenon in the dissipation energy of surrounding rock during the process of gradually increasing displacement. A novel method for determining the optimum supporting opportunity of roadway based on energy dissipation was proposed, based on the above finding. This study can provide an original idea for the determination of roadway supporting opportunity.