This study aims to enhance roadway support effectiveness and construction efficiency. The study proposes a synergistic technology integrating roof grouting with active support techniques in stratified mining environments, enhanced by deep reinforcement learning (DRL) algorithms that adaptively optimize grouting parameters in real time. A mechanical model of a roadway in stratified mining is established, along with a grouting parameter optimization framework based on Deep Deterministic Policy Gradient (DDPG). Using the 20,701 transport roadway in a deep coal mine as the test area, a comparative experimental system is constructed comprising traditional support (Section A), dual-layer grouting and bolting support (DLGBS) (Section B), and DDPG-optimized intelligent support (Section C). Through the combined field monitoring and numerical simulation, the performance differences among the three schemes are systematically evaluated in surrounding rock deformation control, support cost, construction efficiency, and long-term stability. The double-layer grouting technology establishes a multi-level bearing system through shallow grouting for fracture filling and deep anchoring for constraint. In contrast, the DDPG algorithm dynamically adjusts grouting pressure, slurry ratio, and injection rate based on multi-source sensor data. The field monitoring results show that after 30 days of support construction, the traditional support, DLGBS and the roof subsidence of DDPG-optimized support are 45.6, 28.4, and 18.7 millimeters (mm), respectively. The convergence of sidewalls is 32.1, 19.8 and 12.3 mm, respectively, and the floor heave is 12.4, 5.4 and 3.1 mm. Compared with the traditional support, the roof subsidence under DDPG optimization conditions is reduced by about 59.0%. These results are applicable to the 20,701 transport roadway and coal roadway with similar burial depth, coal seam thickness, stratified mining mode and stress conditions, and can provide a reference for grouting and active support design under similar engineering conditions.
No takes yet. Share an insight, caveat, or question.
Wang et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: