基于支持向量机的岩爆模式识别及预测

Pattern recognition and prediction of rockburst based on support vector machine

采用支持向量机,运用分析国内、外工程岩爆数据,以岩石单轴抗压强度与单轴抗拉强度的比值、洞室围岩的最大切向应力与岩石单轴抗压强度比值及弹性能量指数作为评判指标,对典型岩爆进行模式识别即分类,并进行了预测.试验结果表明,该预测方法具有较高的准确率,较好地解决了小样本及非线性等实际问题.

Based on the analysis of domestic and international projects data, a typical rockburst was predicted with support vector machine. The ratio of uniaxial compressive strength to uniaxial tensile strength of rock, the ratio of maximum tangent stress of adjoining rock to uniaxial compressive strength of rock, elastic energy index of rock were selected as a judge of indicators. The results show that this method is reliable and promising, and the model is very useful to solve the problems such as small sample and nonlinearity.