Sign in Subscribe
Periodical publishing

Researchers Develop AI Model for Early Warning of Rock Burst in Coal Mines

Strategic Insights

— 1 min read

Researchers from Xi'an University of Science and Technology have developed a time risk level early warning model that can predict and prevent rock bursts in coal mines. According to the study, the model combines Bayesian optimization algorithm with long-term and short-term memory network, allowing it to accurately warn most "strong" level events with an accuracy of 84.8%. The model has been tested in engineering practice and has shown promising results, providing technical support for rock burst mine monitoring and disaster prevention and control.

Key Takeaways:

  • Researchers from Xi'an University of Science and Technology have developed a time risk level early warning model for rock burst in coal mines.
  • The model combines Bayesian optimization algorithm with long-term and short-term memory network, utilizing machine learning techniques.
  • The model has been tested in engineering practice and has shown an accuracy of 84.8% in predicting and preventing rock bursts.
  • The model's early warning indicators include B value, A value, A(b) value, lack of earthquake, d F, S and H(t).
  • The model has good applicability and practicability in the time early warning of rock burst, providing technical support for rock burst mine monitoring and disaster prevention and control.
  • A research article detailing the study was published in the Meitan xuebao, a journal published by the Editorial Office of Journal of China Coal Society.
  • The study's authors include Feng CUI, Shifeng HE, Zhong LUO, Cheng ZONG, Haodang LI, Liqiang MA, Zhipeng ZHAO, and Xu YANG.

Statistics:

  • The accuracy of the test set is 84.8%.
  • The model has been tested in the Kuangou Coal Mine, where the main types of rock burst occurred were analyzed by means of field investigation, theoretical analysis, and machine learning.
  • The breakage zone under the coal pillar has a focal energy greater than other areas.
  • The energy and frequency under goaf are obviously low.
  • The established model realizes the graded early warning of daily impact risk.

Sources:

  • Research on multi-index early warning of rock burst based on bayesian optimization algorithm and machine learning. Meitan xuebao, 2025, 50(S1):297-313.
  • Editorial Office of Journal of China Coal Society, publisher of Meitan xuebao.
  • Online version of the journal article available at: https://doi-org.sdpl.idm.oclc.org/10.13225/j.cnki.jccs.2024.1499

Read more