Xiaowan Lei

Accurate mine pressure prediction is essential for understanding roof–support interaction and supporting intelligent monitoring in fully mechanized longwall mining. In underground production, hydraulic-support pressure sensors provide continuous pressure sequences that reflect the mechanical response of the support–roof system. However, these sequences are affected by local mining disturbances, missing records, abnormal zero-value segments, nonstationary fluctuations, and periodic weighting, which make future pressure forecasting challenging. To address this issue, an LSTM–Transformer hybrid model is proposed for hydraulic-support pressure forecasting. The LSTM module extracts local nonlinear pressure-evolution features from recent historical windows, whereas the Transformer module captures temporal dependencies and periodic pressure patterns through global sequence modeling. Support-wise experiments were conducted using field monitoring data from Yili No. 1 Mine, and the pressure sequence of each support was processed independently to avoid mixing information from different support locations. In the representative test case, the proposed model achieved an R² of 0.971 and reduced the MAE to 0.471 MPa, while improving the phase consistency of predicted pressure peaks. Further analysis indicates that sufficient historical data coverage is necessary to capture complete pressure-evolution cycles, and that the 25-step forecasting case maintains stable accuracy for short-term mine pressure estimation. These findings demonstrate the feasibility of the proposed approach for hydraulic-support pressure prediction under the monitored conditions of the studied working face.

Hydraulic support (Hs) is an important support equipment in coal mining. With the continuous increase in coal mining intensity, stricter technical specifications have been put forward for the effectiveness of Hs. Hs is always in a dynamic coupling state with the surrounding rock. Investigating its dynamic adaptation characteristics in relation to surrounding rock is of great significance for improving its performance. In this work, a numerical analysis model of the ‘support–surrounding rock’ coupling system was established by taking the Hs (type ZZ 17000/33/72D) in the Kouzidong coal mine as an example and using explicit dynamic analysis software Ls-dyna R12. The dynamic response and pressure distribution characteristics of the hydraulic ‘support–surrounding rock’ coupling system under different mining heights and load conditions were investigated. The key vulnerable connection components of the Hs and their critical connection unit instability conditions were identified. These findings provide a theoretical basis for the structural optimization of four-column Hs, which will be beneficial in promoting the improvement of its load-bearing stability.

As key supporting equipment in coal mining, hydraulic supports are vulnerable to impact pressure from roof movement and deformation. In this paper, a mechanical-hydraulic co-simulation platform for hydraulic supports is established. Moreover, the rationality of the simulation platform is verified. Based on this platform, the rigid-flexible coupling impact dynamics model of hydraulic support is built. Finally, by delaying the opening time of the relief valve, the energy dissipation problem of the relief valve hysteresis effect on the hydraulic support system under the rotary impact is discussed. The results indicate that the rotary load acting on the hydraulic support decreases gradually with the backward movement of the roof rotary position, which causes the peak pressure in the column to decrease (by 69 MPa). The hinge point load of different parts shows different load transfer laws. The hysteresis effect of the relief valve prolongs the energy release time of the system, increasing the pressure in the column by 23 MPa. The instantaneous opening speed of the relief valve spool reaches 15.7 m/s, and the hinge point between the top beam and the column is most sensitive to the hysteresis effect (impact coefficient increases by 0.63).