⚡ electrical engineering

Mechanism of Fault-Related Water Inrush during Roadway Excavation in Deep Phosphate Mines and Construction of a Predictive Model for the Safe Thickness of Water-Resistant Rock Mass

This study elucidates the mechanism of fault-induced water inrush in deep phosphate mines through fluid–solid coupling simulations and establishes a high-accuracy predictive model for safe water-resistant rock thickness based on orthogonal design and multiple linear regression.

Original authors: CHEN JUNGUO, LI BO, ZHANG Bei Bei, LI CHAO YI, CHEN WEI, LIU Shou QIANG, LI TAO

Published 2026-10-19

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: CHEN JUNGUO, LI BO, ZHANG Bei Bei, LI CHAO YI, CHEN WEI, LIU Shou QIANG, LI TAO

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Deep underground, where the earth presses down with immense weight and water waits in hidden pockets under high pressure, mining is a delicate balancing act. When engineers dig tunnels, or roadways, through rock, they disturb a natural state of equilibrium. The rock around the tunnel tries to settle into the new empty space, while the water trapped in cracks and pores pushes back. In places where the ground is fractured by faults—giant cracks in the earth's crust—this interaction can become dangerous. If the rock barrier between the tunnel and a water-filled fault is too thin, the pressure can force water to burst through, flooding the mine and threatening the lives of those below. This is not just a problem for coal miners; it is a critical challenge for phosphate mines, which are digging deeper to find the rock needed for global food production. Understanding exactly how and when this water breakthrough happens, and knowing precisely how thick the rock barrier must be to stop it, is a matter of safety and survival.

Researchers from several universities in China, led by Junguo Chen and Bo Li, turned their attention to the Yingping phosphate mine in central Guizhou. This area is geologically complex, filled with karst landscapes where water dissolves rock to create caves, and intersected by faults that act as highways for groundwater. The team focused on a specific fault, known as F102, which runs near a roadway being excavated at a depth of 830 meters. To understand the invisible forces at play, they built a detailed digital replica of the mine and the surrounding rock layers using powerful computer software. This simulation allowed them to watch, in slow motion, how the rock and water reacted as the tunnel advanced step by step toward the fault. They did not just look at the rock; they tracked how stress built up, how the rock cracked and deformed, and how the pressure of the water changed, creating a complete picture of the disaster waiting to happen.

The simulation revealed a clear and terrifying sequence of events. As the tunnel dug closer to the fault, the rock around the tunnel began to crack and weaken, forming a zone of damage that grew outward like a wedge. Initially, this damage was small and stayed close to the tunnel walls. But as the tunnel face got within about 20 meters of the fault, the wedge of broken rock expanded rapidly, reaching out to touch the fault itself. Once this connection was made, a continuous path opened up from the high-pressure water source, through the fault, through the damaged rock, and directly into the tunnel. The moment this path formed, the water pressure in the fault dropped sharply, signaling that the water was rushing into the mine. The researchers observed that the roof of the tunnel began to sink and the floor to heave upward, while the rock ahead of the tunnel face became intensely compressed. These physical changes, combined with the sudden drop in water pressure, served as the warning signs that a flood was imminent.

To find out exactly how thick the rock barrier needed to be to prevent this, the team ran hundreds of computer experiments, changing one variable at a time. They tested different angles of the fault, different sizes of the tunnel, different depths of the mine, different strengths of the rock, and, most importantly, different levels of water pressure. By analyzing the results, they discovered that not all factors were equally important. While the size of the tunnel and the angle of the fault mattered, the single most critical factor was the pressure of the water itself. The higher the water pressure, the thicker the rock barrier needed to be. The stiffness of the rock also played a major role; harder, stiffer rock required a thinner barrier than softer rock. They found that the depth of the mine and the size of the tunnel had a moderate effect, while the angle of the fault had the least influence on the required safety margin.

Using these findings, the researchers created a new mathematical formula to predict the safe thickness of the rock barrier. Unlike older rules of thumb that were designed for coal mines in shallower, simpler conditions, this new model was built specifically for the deep, high-pressure environment of phosphate mining. When they tested their formula against the actual conditions at the Yingping mine, it proved highly accurate. The model predicted that a rock barrier of about 17 meters would be safe, while the actual mine experienced a water inrush when the barrier was reduced to 20 meters. This means the model was correct within a margin of error of less than 15 percent, a significant improvement over existing methods which would have suggested a barrier of only about 7 meters—far too thin to stop the flood. The study confirms that in deep mining, relying on old, simple formulas can be dangerous, and that a precise understanding of the water pressure and rock strength is essential to keep the tunnels dry and the miners safe.

Technical Summary: Mechanism of Fault-Related Water Inrush and Safe Thickness Prediction in Deep Phosphate Mines

Problem Statement
Fault-induced water inrush represents a critical safety hazard during roadway excavation in deep phosphate mines, particularly in the karst-fault composite hydrogeological systems of central Guizhou. Unlike coal mines, deep phosphate mining involves large cross-section roadways, high in-situ stress, high water pressure, and complex geological structures that render existing coal-mine empirical formulas inadequate. The specific challenge addressed is the lack of a systematic understanding of the disaster-causing mechanism involving surrounding rock failure and fault activation under excavation disturbance, as well as the absence of an accurate method to predict the safe thickness of water-resistant rock pillars for these specific conditions.

Methodology
The study utilized the F102 water-conducting fault at the Yingping phosphate mine as the engineering background. The research employed a multi-stage approach:

Numerical Simulation: A three-dimensional fluid–solid coupling model was constructed in COMSOL Multiphysics based on Biot's poroelasticity theory and the Brinkman–Forchheimer equation (to characterize nonlinear seepage in the fault zone). The model simulated the spatiotemporal evolution of stress, displacement, plastic zones, and pore water pressure during roadway excavation approaching the fault.
Orthogonal Experimental Design: An $L_{25}(5^5)$ orthogonal experiment was conducted to quantify the influence of five key factors: fault dip angle, roadway size, fault water pressure, burial depth, and surrounding rock elastic modulus.
Statistical Analysis: Range analysis and Analysis of Variance (ANOVA) were used to determine the significance and weight of each factor.
Predictive Modeling: Multiple linear regression was applied to the simulation data to develop a mathematical model for predicting the safe thickness of the water-resistant rock mass.
Validation: The proposed model was validated against field monitoring data from the Yingping mine and compared against traditional empirical formulas from the Detailed Rules for Mine Water Prevention and Control in Coal Mines.

Key Results

Evolution Mechanism: The study revealed that water inrush is a multi-field coupled process. As excavation approaches the fault, the plastic zone expands in a wedge shape from the roadway periphery toward the fault. At a critical distance (70 m in the case study), the plastic zone connects with the fault fracture zone, forming a through-going "roadway–damage zone–fault" pathway. Precursory characteristics include stress concentration, roof subsidence, floor heave, and a sharp drop in pore water pressure (observed reductions of 46–53% in monitoring points).
Factor Significance: Sensitivity analysis established the ranking of influencing factors on safe thickness as: Fault water pressure > Elastic modulus > Burial depth > Roadway size > Fault dip angle. Statistical analysis (ANOVA) confirmed that only fault water pressure has a statistically significant effect ( $F > F_{critical}$ ) at the 0.05 significance level, while other factors were not statistically significant in this specific experimental design, though they remain influential in the regression model.
Prediction Model: A multiple linear regression model was derived with a coefficient of determination ( $R^2$ ) of 0.994. The formula is:
$T = -0.0307\theta + 0.3308H + 5.0691P + 0.0075B - 0.0501E + 0.6995$
Where $T$ is the safe thickness, $\theta$ is the dip angle, $H$ is roadway size, $P$ is water pressure, $B$ is burial depth, and $E$ is elastic modulus.
Validation Accuracy: Field validation at the 830 m level crosscut showed the model predicted a safe thickness of 17.05 m, while the actual critical thickness observed was 20 m. This resulted in a prediction accuracy of 85.25%. In contrast, the traditional coal-mine empirical formula yielded a thickness of only 6.89 m, significantly underestimating the required safety margin.

Significance and Contributions
The paper claims to provide a theoretical and technical basis for preventing fault-induced water inrush in deep phosphate mines, which differ fundamentally from coal mining environments. The primary contributions are:

Mechanistic Insight: It elucidates the specific fluid–solid coupling evolution of fault-induced water inrush in deep phosphate settings, identifying the wedge-shaped plastic zone expansion and pore pressure drop as key triggers.
Methodological Shift: The study advocates for a shift from "empirical application" (relying on coal-mine codes) to a "mechanism-driven, scenario-adaptive" approach. The proposed regression model significantly outperforms traditional empirical formulas in accuracy and applicability to deep, high-stress, high-pressure environments.
Engineering Application: The validated predictive model offers a more reliable tool for designing waterproof rock pillars, potentially preventing catastrophic water inrush accidents that traditional methods might fail to avert due to underestimation of required thickness.

Limitations
The authors acknowledge that the numerical model simplifies actual hydrogeological conditions and does not fully account for the strong nonlinearity of fluid–solid coupling or the influence of initial rock fractures. Future work is suggested to incorporate three-dimensional fracture reconstruction and more comprehensive interaction analyses among control variables.

More like this