TL;DR
This paper investigates how fractures evolve in the rock surrounding underground coal gasification (UCG) cavities. By developing numerical models, the researchers identified key geological and operational factors that control fracture propagation, leading to a new predictive model for fracture height and depth, crucial for UCG safety assessments.
What Problem Does This Paper Address?
Underground Coal Gasification (UCG) offers a promising method for clean coal utilization, but its large-scale implementation is hindered by safety concerns, particularly regarding fracture propagation in the surrounding rock. These fractures can lead to gas emissions and water leakage, posing environmental risks and impacting the efficiency of the gasification process. While previous research has explored fracture characteristics in UCG, a comprehensive understanding of the evolution of fracture zones and the precise influence of various factors, especially in the context of cavity expansion, remained unclear. This paper aims to fill this gap by systematically analyzing the mechanisms of fracture propagation and developing a predictive model for fracture dimensions.
Methods
The researchers employed a multi-faceted approach:
- Numerical Modeling: The core of the study involved developing and utilizing a 2D axisymmetric numerical model in ABAQUS software. This model simulated the thermal-mechanical coupling effects during cavity expansion, considering various geological (rock grade, coal seam thickness) and operational (gasification pressure, cavity dimensions) conditions.
- Factor Analysis: A systematic analysis was conducted by varying six key influence factors (rock grade, cavity depth, coal grade, cavity width-to-height ratio, coal seam thickness, and gasification pressure) across five levels each. This allowed for the quantification of their impact on fracture evolution.
- Mechanism Investigation: The study delved into the physical mechanisms of rock breakage, distinguishing between shear and tensile fractures and observing their progression from initial near-field breakage to roof caving and eventual baseboard fracturing.
- Predictive Model Development: Based on the observed linear relationships between influence factors and fracture dimensions, a model was proposed to predict the fracture height in the roof and the fracture depth in the baseboard. This model incorporates the maximum of tensile and shear fracture contributions and considers thermal damage limitations.
- Model Validation: The developed prediction model was validated using an orthogonal test design. This involved performing numerical simulations for 25 different case combinations and comparing the model's predictions against these simulation results.
Datasets
The study did not introduce new benchmark datasets. Instead, it relied on:
- Experimental Data: Mechanical parameters of coal and sandstone under varying temperatures were referenced from existing laboratory tests (e.g., Otto and Kempka [11], Liu et al. [49]).
- Numerical Simulation Results: The primary "data" for model development and validation came from the numerical simulations conducted by the authors themselves, based on the defined model and parameters. An orthogonal test matrix was used to systematically generate these simulation results for validation.
Key Insights and Ideas
- Cavity Expansion Stages: The process of cavity expansion is not uniform. It begins with near-field rock breakage, followed by the formation of a caving arch in the roof, and then the propagation of shear bands.
- Dominant Fracture Types: In the roof layers, both shear and tensile fractures contribute to the overall fracture zone. However, the fracture depth in the baseboard is primarily driven by tensile fracture.
- Critical Influence Factors: The grade of confining rock, overburden pressure, cavity dimensions, and gasification pressure are identified as the most significant factors controlling fracture evolution, particularly the fracture height in the roof.
- Linear Relationships: A key finding is the nearly linear relationship between these influence factors and the resulting fracture height and depth, which forms the basis for the predictive model.
- Thermal Damage Limit: The thermal damage to the confining rock, while present, has a limited depth (around 3.5 m), acting as a constraint on the maximum fracture depth.
- Predictive Model Utility: The proposed model, validated against orthogonal test results, demonstrates a good capacity for predicting fracture distribution, offering a valuable tool for safety assessments.
Key Results
- The study successfully simulated the evolution of fracture zones during cavity expansion, showing the transition from initial breakage to roof caving and shear band propagation.
- Key influence factors were identified, with their impact on fracture height and depth quantified. For instance, higher confining rock grade generally decreases shear fracture height, while increased overburden pressure, cavity width-to-height ratio, and coal seam thickness tend to increase shear fracture height.
- A predictive model was established that, when validated, showed a similar tendency to the orthogonal test results for both fracture height and depth. The prediction results for fracture height were slightly higher than the test data, suggesting a potentially safer estimation for fluid channel assessment.
Limitations and Future Work
- Model Simplification: The numerical model is a 2D axisymmetric representation, which simplifies the complex 3D reality of UCG cavities.
- Data Fluctuations: While the model shows good accuracy, some fluctuations in the data, particularly for fracture depth, were observed, leading to slightly larger errors in regression.
- Field Verification: The authors emphasize the need for field verification of the proposed mechanism and model.
- Water/Gas Emission Assessment: Further research should focus on the practical application of the model for assessing water leakage and gas emission through the predicted fracture zones.
- Complex Geological Conditions: The study primarily focuses on specific geological and operating conditions. Future work could explore a wider range of these parameters and their interactions.
Summary updated