Risk Prediction of Coal and Gas Outburst in Deep Coal Mines Based on the SAPSO-ELM Algorithm

Area of Science:

• Mining Engineering
• Geological Engineering
• Computational Science

Background:

• Coal-gas outbursts pose significant risks in deep mining operations.
• Current prediction methods lack accuracy and efficiency.
• Effective risk management is crucial for preventing accidents and casualties.

Purpose of the Study:

• To develop a highly accurate and efficient prediction method for coal-gas outbursts in deep coal mines.
• To improve the safety management capabilities of coal mine enterprises.
• To address the limitations of existing coal-gas outburst prediction techniques.

Main Methods:

• Kernel Principal Component Analysis (KPCA) for high-dimensional nonlinear data processing.
• Simulated Annealing Particle Swarm Optimization (SAPSO) to optimize the Extreme Learning Machine (ELM) parameters.
• Development of a SAPSO-ELM based risk prediction model for deep coal mines.

Main Results:

• The SAPSO-ELM algorithm significantly enhanced the accuracy of coal-gas outburst risk prediction.
• Achieved a prediction accuracy rate of 100%, outperforming standard ELM and PSO-ELM algorithms.
• Demonstrated the effectiveness of the proposed method in real-world deep coal mine scenarios.

Conclusions:

• The SAPSO-ELM model offers a superior approach for predicting coal-gas outbursts.
• This research contributes to the theoretical and practical aspects of safety management in deep coal mines.
• The developed method empowers coal mine enterprises to better manage outburst risks and enhance operational safety.