A study on drum cutting properties with full-scale experiments and numerical simulations

Xuefeng Li, Shibo Wang, Shirong Ge, Reza Malekian, Zhixiong Li, Yifei LiMeasurement 2017 · DOI

TL;DR

This paper bridges the gap between laboratory-scale experiments and real-world applications in underground coal mining by developing and validating a full-scale numerical model for shearer drum cutting. The study demonstrates that a discrete element method (DEM) simulation using PFC3D can accurately predict drum cutting loads, specific energy, and torque fluctuations, offering a more accessible and reliable alternative to costly physical testing for shearer design.

What Problem Does This Paper Address?

Underground coal mining heavily relies on longwall mining with shearers, where the cutting performance of the shearer drum is crucial for productivity and reliability. Accurately predicting the cutting loads on the drum is essential for designing efficient and robust shearers. While theoretical models and small-scale experimental setups exist, they often struggle to capture the complexities of real-world conditions due to scale differences, boundary effects, and material anisotropy. Full-scale experiments are the most reliable but are prohibitively expensive and difficult to conduct. This paper addresses the need for a cost-effective and accurate method to predict shearer drum cutting loads by developing and validating a numerical simulation model.

Methods

The study employs a two-pronged approach:

  • Full-Scale Experimental Testing: Researchers conducted full-scale cutting tests using a real shearer in a simulated underground coal mining environment. They measured pick forces and drum torque using embedded strain sensors and a force sensor on the idler shaft, respectively. Tests were performed at various traction speeds (1.5, 3, and 5 m/min) while maintaining a constant drum rotation speed (28 rpm).
  • Numerical Simulation with PFC3D: A 3D discrete element method (DEM) model was developed using Particle Flow Code in three dimensions (PFC3D). This approach models the coal mass as an assembly of discrete particles, allowing for the simulation of rock fracture and chip formation. The model was calibrated using uniaxial compressive strength (UCS) tests on artificial coal samples. The simulation replicated the full-scale cutting tests, modeling both the front and back drums under identical conditions. The parallel bond contact model was used for the coal particles.

Datasets

The study utilized experimental data from full-scale shearer drum cutting tests conducted at the National Energy Coal Mining Machinery Equipment Research and Design Center of China. Additionally, mechanical properties of artificial coal samples (UCS, Young's modulus, Poisson's ratio, and Brazilian tensile strength) were measured and used for calibrating the PFC3D model. No new benchmark datasets were introduced; rather, the experimental data served as the ground truth for validating the simulation.

Key Insights and Ideas

  • Bridging the Scale Gap: The study successfully demonstrates that a DEM model, when properly calibrated, can accurately represent the complex cutting process of a full-scale shearer drum, overcoming the limitations of small-scale experiments.
  • Traction Speed Dominance: Both experimental and simulation results consistently show that increasing traction speed leads to a linear increase in mean drum torque and its standard deviation. This highlights traction speed as a critical parameter influencing cutting loads.
  • Torque Fluctuation Behavior: The torque fluctuation coefficient decreases with increasing traction speed, indicating more stable torque despite higher overall loads at higher speeds. This is a crucial insight for shearer control and vibration management.
  • Frequency Domain Insights: Analysis of drum torque in the frequency domain revealed characteristic frequencies below 50 Hz, consistent between experiments and simulations. This suggests that PFC3D can effectively capture the dynamic behavior of the cutting process.
  • PFC3D as a Viable Tool: The study strongly advocates for PFC3D as an easier, faster, and reliable method for predicting drum cutting loads and aiding in shearer drum design, especially given the difficulties and costs associated with full-scale testing.

Key Results

  • Torque Prediction: The simulation results for mean drum torque and its standard deviation showed reasonable agreement with experimental data, with high correlation coefficients (R² ≥ 0.8825 for mean torque, R² ≥ 0.9437 for standard deviation).
  • Specific Energy: The specific energy (energy consumed per unit volume of coal cut) increased with traction speed in both experiments and simulations, with a high correlation coefficient (R² ≥ 0.9998 for simulation).
  • Torque Fluctuation: The simulation accurately reproduced the trend of decreasing torque fluctuation coefficient with increasing traction speed, as observed in experiments.
  • Frequency Analysis: The PSD analysis of the torque signals confirmed that the vibration of the drum torque is composed of multiple frequencies below 50 Hz, and these characteristic frequencies were well-reproduced by the PFC3D model.

Limitations and Future Work

  • Rock Anisotropy: The authors acknowledge that the difference between experimental and simulated drum torque and specific energy might be partly due to the effect of rock anisotropy, which is more complex to fully capture in the current DEM model compared to the homogeneous artificial coal mass.
  • Simulation Time: For low traction speeds, the simulation time was noted as potentially insufficient to fully capture the statistical chip volume, leading to some discrepancies in specific energy calculations. The authors suggest longer simulation durations for such conditions.
  • Future Work: The paper suggests future investigations into the effects of other kinematic parameters (e.g., rotation speed) and structural parameters (e.g., pick arrangement, number and dip angle of spiral vanes) on cutting and loading efficiency using the developed PFC3D model. This would further enhance the utility of the model for comprehensive shearer design.

Summary updated

BibTeX
@inproceedings{li2017study,
  author = {Xuefeng Li and Shibo Wang and Shirong Ge and Reza Malekian and Zhixiong Li and Yifei Li},
  title = {{A study on drum cutting properties with full-scale experiments and numerical simulations}},
  booktitle = {Measurement},
  year = {2017},
  doi = {10.1016/j.measurement.2017.09.006},
}
Figure