During the mining process of Hami open-pit coal mine, a large number of cracks appeared on the ground surface near the top of the end slope, which affected the stability of the slope and the subsequent safe mining of resources. By analyzing the deformation of rock stratum, the position of weak interlayer in the slope was determined. Considering the rock mass with weak interlayer as a complete mechanical system, the mechanical model of the rock mass system with weak interlayer was established, and the evaluation index of rock failure degree before slope instability was proposed. The failure mode of slope with weak interlayer is as follows: sliding failure-cutting layer failure-tension failure. The safety factor of slope decreases first and then increases with the increase of the depth of the weak interlayer, and gradually decreases with the increase of the inclination of the weak interlayer. The pressing foot of slope can be used in landslide control of slope with weak interlayers, and the optimal pressing foot solution to ensure the safe mining of final slope coal resources was determined to be a pressing foot height of 35 m and pressing foot width of 20 m.

In this study, a detailed analysis was conducted to evaluate the impacts of the deviatoric stress component and spherical stress component on the stability of surrounding rocks in the roadway via the theoretical analysis and calculation and numerical simulation. Based on the analysis, the distribution laws guiding the main stress differences, plastic zone, convergence of surrounding rocks, and third invariant of stress under various conditions (such as equal spherical stress and unequal deviatoric stress and equal deviatoric stress and unequal spherical stress) were developed, providing an optimization scheme for roadway support misunderstanding under the conditions of high spherical stress field and high deviator stress field. The study further reveals that under the circumstance of the constant spherical stress, the greater the deviatoric stress, the plastic zone range of the surrounding rock of the roadway, the range of tensile deformation of the surrounding rock, the amount of convergence of the surrounding rock, the probability of separation of the roof and floor of the roadway, and the principal stress difference and the main stress, the greater the concentration range of the maximum stress difference is, and the maximum principal stress difference is mainly concentrated in the roof and floor rocks of the roadway, and the greater the deviatoric stress, the greater the probability that the roof and floor rocks of the roadway will be separated, and the maximum principal stress difference is mainly concentrated in the roof and floor rocks of the roadway, the greater the deviator stress, the greater the concentration range of the maximum value of the principal stress difference and the principal stress difference; when the deviator stress is constant, the range of the plastic zone and the maximum principal stress difference concentration range of the surrounding rock of the roadway decrease with the increase of the ball stress, and the principal stress difference, the amount of convergence of the surrounding rock, and the range of tensile deformation increase with the increase of the ball stress. The maximum principal stress difference is mainly concentrated in the roof and floor rocks of the roadway. The principal stress difference increases with the increase of the spherical stress, and the maximum concentration range of the principal stress difference decreases with the increase of the spherical stress. After the method proposed in this paper optimizes the actual roadway support on site, the surrounding rock deformation of the roadway is small and the control is relatively ideal, which basically meets the engineering needs.

Based on the variation range of the stress lode angle, the in situ rock stress is divided into σ_v‐type stress field, σ_H‐type stress field, and σ_h‐type stress field. Through theoretical analysis, the principal stress difference distribution law and plastic zone distribution pattern around the roadway in different types of stress fields are obtained. Theoretical and numerical simulation calculation results show that under different stress lode angle conditions, the principal stress difference distribution of the surrounding rock of the roadway is greatly different, which has a direct effect on the shape and range of the plastic zone of the surrounding rock of the roadway. In the σ_v‐type stress field and the σ_H‐type stress field, the shape of the plastic zone of the roadway surrounding rock is mainly oval and “butterfly,” while in the σ_h‐type stress field, the shape of the plastic zone of the roadway surrounding rock is mainly oval. The laboratory test proves that the stress gradient has an important effect on the damage degree of the surrounding rock of the roadway. The larger the stress gradient, the higher the strength of the rock mass and the more severe the damage. The change of the stress lode angle will affect the distribution law of the stress gradient of the surrounding rock of the roadway, thus affecting the degree of fragmentation of the surrounding rock. In type σ_v and type σ_H stress fields, the surrounding rock of the shoulder can be regarded as a key part of the roadway. In the σ_h‐type stress field, the plastic zones of the surrounding rocks of the roadway are more evenly distributed, and the damage range is less affected by θ. The influence law of the stress lode angle on the stability of the roadway has been well verified by field observation, and effective support measures have been proposed.

The feature extraction of high‐precision microseismic signals is an important prerequisite for multicategory recognition of microseismic signals, and it is also an important basis for intelligent sensing modules in smart mines. Aiming at the problem of unobvious feature extraction of multiclass mine microseismic signals, this paper is based on the unsupervised learning method in the deep learning method, combined with wavelet packet energy ratio and empirical modulus singular value decomposition, and proposes a method based on wavelet packet energy and empirical modulus singular value decomposition and proposes a method (M‐W&E) based on wavelet packet energy and empirical modulus singular value decomposition. This method firstly performs empirical modulus singular value decomposition and wavelet packet energy ratio on the microseismic signal to construct the basic feature vector and then uses the unsupervised learning algorithm to perform the unsupervised learning method feature fusion of the basic feature vector to construct the fused feature vector. After visualization by t‐SNE, various distinctions in the fusion feature vector are more obvious. After testing the fusion feature classification using SVM, it is found that the recognition rate of the new feature after feature fusion is better than that of a single wavelet packet empirical energy component and singular value of empirical modulus, which basically meets the engineering needs and is a mine microseism. The signal extraction and feature enhancement fusion of multiclass samples provide a new idea.