中文摘要：

针对矿井涌水量典型的非线性特征,应用相空间重构与支持向量机耦合方法进行预测。将矿井涌水量的时间序列作相空间重构,并以最小嵌入维数作为支持向量机的输入节点,根据支持向量机原理建立矿井涌水量的预测模型。实例计算表明：与其他维数相比,当嵌入维数为4时,矿井涌水量的预测精度最高,说明引入最小嵌入维数是正确的。为检验该方法预测的可靠性,分别采用最小二乘法、指数函数法、相空间重构与支持向量机耦合法对实际矿井涌水量观测值进行回归预测。结果表明,非线性方法的预测效果比线性方法更佳。不同核函数预测结果证实RBF是最优的。

英文摘要：

Because the mine water inrush has typical nonlinear characteristic,the method of coupling with the phase space reconstructed and the support vector machines is used to predict it. The phase space of the mine water inrush time series is reconstructed. And the minimum embedding dimension is used for the input node of the support vector machines. The prediction model of time series is established based on the support vector machines. The example calculations show that its prediction precision is the highest comparing with other dimensions when the minimum embedding dimension is four. It proves that considering the minimum embedding dimension is correct. For verifying the reliability of the purposed method,the regression predictions of the practical water inrush are done with least square method,index fitting method and the method of coupling the phase space reconstructed and the support vector machines respectively. The results show that the prediction effect of the nonlinear method is better than the linear one. It is proved that RBF is the best to predict mine water inrush compared with other kernal functions.