Mechanical Analysis and Optimization of Mine Roadway Support Structure Empowered by Random Optimization Algorithm

The stability of roadway support structures has a significant impact on the economic benefits and personnel production safety of mines. To analyze the stability state of the mining roadway support structure, this study integrates the support anchor rod and rock structure into a stacked support structure for mechanical characteristic analysis. The optimized Back Propagation (BP) network model in the Stochastic Gradient Descent (SGD) algorithm is introduced to predict the range of loosening circle of mine roadway support structure. The input features of the model include rock rebound value, burial depth, degree of joint development, and tunnel span. In the experiment, the absolute value of the prediction error of the research model is in the range of 1.92 cm to 6.10 cm, with the highest error proportion of 4.9%, and the prediction error is lower than that of other models. Based on the prediction results of the loosening circle range, this study optimizes the parameters of the roadway support structure. Before optimization, the maximum settlement of the top rock layer of the Non-Mining Roadway (NMR) and mining roadway is 43.93 mm and 59.81 mm. After optimization, the settlement and maximum settlement of the top rock layer of the NMR and mining roadway decrease to 32.74 mm and 37.66 mm. The experiment shows that the research method can accurately predict the stability of the mine support structure and has a guiding role in optimizing the mine support structure