Backfill mining is widely used to control strata movement and improve the stress environment in China’s coal mines. In the present study, the effects of backfill mining on strata movement and water inrush were studied based on a case study conducted in Caozhuang Coal Mine. The in-situ investigation measured abutment pressure distribution (APD), roof floor displacement (RFD), and vertical stress in the backfill area. Results are as follows: (i) The range and peak of APD, RFD, and vertical stress in the backfill area are smaller than in traditional longwall mining with the caving method. (ii) Backfill mining could change the movement form and amplitude of overburden and improve the ground pressure environment. (iii) Floor failure depth (FFD) is much smaller in backfill mining. Backfill mining can be an effective method for floor water inrush prevention.

The coal mine accidents seriously affect the safety and efficiency of mining for coal mining enterprises. The reliable emergency rescue (ER) processes are explored to minimize the loss of accidents. This paper introduces the stochastic Petri net (SPN) and Markov chain (MC) models based on the system structure flow to analyze the ER processes of coal mine accidents. In addition, a triangle fuzzy strategy is presented to optimize the SPN model. The “9·28” major water inrush accident in Shanxi Fenxi Zhengsheng Coal Company of China is adopted to evaluate the time performance and accident data of the ER process. The MC model-based steady-state probabilities of the system under various states are used to calculate the average delay time of this system. The triangular fuzzy strategy is used to analyze the change value of the total time in the ER system at the unit transition speed when the firing rate of each transition is changed, which finds the most time-consuming key activities in the ER process. The results show that SPN and MC can reflect the dynamic behaviors of ER process, which provides a reference for the rescue operations of other coal mine accidents. The triangular fuzzy strategy can quickly find out the key activities affecting the ER time, which greatly decreases the calculations generated by analyzing the total time of the system changed at the unit transition speed.

In China, soft rock roof makes up the majority of coal mine roof, yet it is easy to break due to low strength and poor integrity. As such, it is difficult for the traditional pillar-based roadway protection means and pillarless mining techniques to effectively control the roadway roof. In order to solve the problems with pillarless mining under soft rock roof conditions, using the 1510 working face of Xinyi Coal Mine as an example, a gob-side caving roadway forming (GSCRF) technique for broken immediate roof is developed. This paper discusses the adaptability and feasibility of this technology through theoretical modeling and on-site engineering testing. A roadway rock surrounding control scheme based on “cable + steel beams + yielding prop” is further designed, and field tests and monitoring are carried out. Field tests show that, during GSCRF of the 1510 working face, the maximum subsidence of the roof remains within 200 mm. The convergence of the two sides causes stabilization. The tension on the anchor cables is gradually becoming stable. The monitoring results show that the roadway has a good stress environment and the surrounding rock is effectively controlled. Compared with the traditional pillarless mining mode, this technology has the technical advantages of achieving complete elimination of coal pillars, reduced pressure on the roadway roof, and interference-free mining of the working face. The research outcome can provide useful reference for pillarless mining by GSCRF and a solution for pillarless mining under soft rock roof conditions.