In order to better solve the phenomenon of low oxygen in the corner of return airway caused by abnormal gas emission in goaf during shallow coal seam mining, by analyzing the source and reason of low oxygen phenomenon, a prediction model of oxygen concentration in the corner of return airway based on genetic algorithm (GA) and random forest (RF) technology was proposed. The training sample set was established by using the field data obtained from actual monitoring, including the oxygen concentration in the return airway corner, the periodic pressure step distance of the roof, the surface temperature and atmospheric pressure. GA was used to optimize the parameters in the RF model, including trees and leaves in the forest. The results showed that the model prediction error was minimum when the number of trees was 398 and the number of leaves was 1. In addition, GA was used to optimize the number of hidden neurons and the initial weight threshold of the back-propagation neural network (BPNN). In order to verify the superiority of the model, the GA optimized RF and BPNN model are compared with the conventional RF and BPNN model. Analyze the average absolute percentage error (MAPE), root mean square error (RMSE), and average absolute error (MAE) of the prediction data of each model. The results show that the optimized RF prediction model is better than other models in terms of prediction accuracy.

The essence of coal-rock dynamic disasters is the instantaneous instability that occurs when the coal-rock combination system reaches its strength limit. The differing mechanical properties of coal and rock lead to mismatching deformation within the combination, thereby altering the mechanical characteristics of the coal-rock body under combination structures. Consequently, studying the mismatching horizontal deformation of coal-rock combinations is significant for the prevention and control of coal-rock dynamic disasters. To investigate the mismatching horizontal deformation characteristics of primary coal-rock combinations, this study conducted uniaxial and conventional triaxial loading experiments on primary coal-rock combinations to obtain their strength and horizontal deformation characteristics. Subsequently, numerical simulations were performed to analyze the factors influencing the horizontal deformation of primary coal-rock combinations, ultimately leading to the construction of a horizontal deformation distribution model for primary coal-rock combinations. The research results indicate that due to the influence of the coal-rock interface, the primary coal-rock combinations exhibit continuous and smooth mismatching horizontal deformation under loading conditions, with the magnitude of deformation ordered as coal > interface > rock. The influence of the interface on the horizontal deformation of the coal-rock body does not suddenly vanish; rather, it gradually diminishes with increasing distance from the interface, becoming negligible at approximately 48% of the model’s lateral dimensions. A horizontal deformation mismatch distribution model for primary coal-rock combinations was established based on the Sigmoid function, and the key parameter b in the model was obtained through fitting with numerical simulation results. The reliability of the model was subsequently validated through experimental results. The findings of this study provide a basis for further research on the mechanical properties and deformation characteristics of coal-rock combinations.