Calculation and program realization of coal pillar setting parameters in Huainan mining area

Area of Science:

• Mining Engineering
• Geotechnical Engineering
• Computational Science

Background:

• Coal pillar retention is vital for mine safety and minimizing ground deformation.
• Current methods face challenges like limited data, long observation times, and high labor costs.
• Accurate determination of protective pillar size and reduction of coal pillar pressure are critical.

Purpose of the Study:

• To develop an automated methodology for coal pillar retention parameter calculation.
• To address challenges in determining optimal pillar size and reducing coal pillar pressure.
• To validate the accuracy of a proposed machine learning model for industrial application.

Main Methods:

• Derived coal pillar retention formulas based on the Three Regulations.
• Integrated total least squares algorithm with surface observation data and MATLAB for automation.
• Developed a predictive model combining a genetic algorithm-optimized ELM neural network and linear regression.

Main Results:

• Automated solution for coal pillar retention parameters using integrated algorithms.
• Established a linear regression model for retention parameters based on geological data.
• Demonstrated that the proposed machine learning algorithm achieves industrial production accuracy.

Conclusions:

• The presented methodology effectively automates coal pillar retention calculations.
• The machine learning approach provides a reliable and accurate solution for mine safety.
• This study offers a practical solution to long-standing challenges in coal pillar management.