A Safety Warning Model Based on IAHA-SVM for Coal Mine Environment

Area of Science:

• Mining Engineering
• Artificial Intelligence
• Environmental Science

Background:

• Coal mining is vital but poses significant underground safety risks.
• Accidents often stem from hazardous underground climate conditions.
• Timely detection of environmental anomalies is crucial for worker safety.

Purpose of the Study:

• To develop an advanced early warning model for coal mine environmental safety.
• To enhance the accuracy and efficiency of safety assessments using AI.
• To improve the detection of underground climate abnormalities.

Main Methods:

• Optimizing Support Vector Machine (SVM) parameters with an Improved Artificial Hummingbird Algorithm (IAHA).
• Incorporating Tent chaos mapping, backward learning, and Levy flight into the IAHA for enhanced performance.
• Utilizing a simplex method to refine the algorithm's iterative process.
• Classifying coal mine safety into four levels based on environmental parameters.

Main Results:

• The IAHA demonstrated improved convergence speed and search accuracy compared to standard algorithms.
• The IAHA-SVM model achieved significantly enhanced performance in classifying coal mine environmental safety.
• The developed model effectively predicts safety levels, aiding in timely interventions.

Conclusions:

• The IAHA-SVM model offers a robust and accurate solution for coal mine safety early warning systems.
• The enhancements to the artificial hummingbird algorithm improve its applicability in complex optimization tasks.
• This approach contributes to mitigating risks and ensuring a safer working environment in underground coal mines.