论文部分内容阅读
根据梧桐庄煤矿涌水量数据,首先用R/S分析计算得到Hurst指数值为0.8627,表明该矿井涌水量变化趋势具有长期相关性和持续性;然后利用混沌时间序列的全域预测法对历年来的数据进行建模预测分析,计算得到了预测模型以及预测值,结果表明,预测数据和实际值误差较小。利用非线性时间序列分析可以揭示矿井涌水的非线性特征,并对其进行预测,从而为矿井涌水量的预测提供了一种新的方法。
According to the data of water inflow from Wutongzhuang coal mine, firstly, the Hurst index is 0.8627 calculated by R / S analysis, which shows that the trend of water inflow in this mine has long-term correlation and persistence. Then, using the global forecasting method of chaotic time series, The data are modeled and predicted, and the prediction model and the predicted value are calculated. The results show that the error between the predicted data and the actual value is small. Nonlinear time series analysis can reveal the nonlinear characteristics of mine gushing water and predict it, which provides a new method for mine water inflow forecasting.