岩爆预测研究现状综述

A Review of the Current Status of Rockburst Prediction Research

摘要: 岩爆是严重威胁深部地下工程安全的关键动力灾害，其精准预测是岩石力学领域的研究热点与难点。文章系统梳理了国内外岩爆预测的研究现状，将现有预测方法归纳为三大类：理论判据预测、现场监测预测及人工智能模型预测。详细阐述了各类方法的核心原理、研究进展与应用实例，并对比分析了其适用条件与局限性。综述表明，理论判据是基础但普适性不足，声发射与微震等监测技术能实现动态预警但易受干扰，而机器学习等智能模型虽预测性能优异但其“黑箱”特性导致可解释性较差。最后，展望了未来岩爆预测向多源信息融合、智能化与精准化发展的趋势，为深化岩爆机理研究与提升灾害防控水平提供了系统性参考。

Abstract: Rockburst is a critical dynamic disaster that severely threatens the safety of deep underground engineering, and its accurate prediction remains a major research focus and challenge in rock mechanics. This paper systematically reviews the current state of rockburst prediction research, categorizing existing methods into three main types: theoretical criterion-based prediction, on-site monitoring-based prediction, and artificial intelligence model-based prediction. It elaborates on the core principles, research progress, and application cases of each method, while providing a comparative analysis of their applicability and limitations. The review indicates that theoretical criteria, while fundamental, often lack universality; monitoring techniques like acoustic emission and microseismic enable dynamic early warning but are susceptible to interference; and intelligent models such as machine learning show superior predictive performance but suffer from poor interpretability due to their “black-box” nature. Finally, the paper discusses future trends, pointing towards the integration of multi-source information, intelligent systems, and precision in rockburst prediction, offering a systematic reference for advancing mechanistic studies and improving disaster prevention and control capabilities.